[go: up one dir, main page]

CN111126432A - A Human Body Type Classification Method for Clothing Design - Google Patents

A Human Body Type Classification Method for Clothing Design Download PDF

Info

Publication number
CN111126432A
CN111126432A CN201911110639.4A CN201911110639A CN111126432A CN 111126432 A CN111126432 A CN 111126432A CN 201911110639 A CN201911110639 A CN 201911110639A CN 111126432 A CN111126432 A CN 111126432A
Authority
CN
China
Prior art keywords
human body
classification
evaluated
fuzzy
category
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN201911110639.4A
Other languages
Chinese (zh)
Other versions
CN111126432B (en
Inventor
董敏
张俊杰
金红淑
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Wuhan Textile University
Original Assignee
Wuhan Textile University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Wuhan Textile University filed Critical Wuhan Textile University
Priority to CN201911110639.4A priority Critical patent/CN111126432B/en
Publication of CN111126432A publication Critical patent/CN111126432A/en
Application granted granted Critical
Publication of CN111126432B publication Critical patent/CN111126432B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/30—Computing systems specially adapted for manufacturing

Landscapes

  • Engineering & Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
  • Image Analysis (AREA)

Abstract

The invention relates to a human body type classification method for clothing design, which comprises the following steps: 1. determining the size of a key body part of a human body, and measuring the size of the key body part of the human body to be evaluated, wherein the size of each part is a characteristic index value; 2. comparing the picture of the body type of the human body to be evaluated with picture data in a body type classification database, and performing category evaluation on the key body part size of the body type of the human body to be evaluated to obtain a preliminary evaluation category of each key body part size; 3. establishing a decision table according to the characteristic index values and the preliminary classification categories, and establishing a fuzzy classification model of the body according to the decision table; 4. and classifying the body types to be evaluated according to the fuzzy classification model and the classification algorithm of the body. The invention establishes a decision table for human body type classification based on the obtained fuzzy data and database comparison data. The invention can classify the body types of human bodies and is used as reference for designing clothes.

Description

Human body type classification method for clothing design
Technical Field
The invention belongs to a related method for customizing clothes, and particularly relates to a human body type classification method and a human body type classification system for clothes design.
Background
In garment design and large-scale customization, human body size analysis is particularly important for meeting personalized requirements of target people. By classifying the body types, more accurate and personalized ready-made garment products (e.g., tops, pants, skirts, etc.) can be designed and produced. The consumption concept centered on human will be developed and popularized in more and more consumption stimulation and market cultivation. Current garment customization is limited to the selection of designs on a limited variety, a certain style, but is still not a fully personalized personal customization in the complete sense. Under the promotion of the technology, the customization mode from style design, material selection, edition opening to finished product delivery is completely designed by the customer based on the customer idea in the future, and the customization mode has a large space.
Body type classification is mainly based on human body size. Currently, information about the size of the human body is obtained by anthropometric measurement. In anthropometric surveys, body dimensions can be measured at many different locations for each person, yielding thousands of data points that should be further analyzed to determine important dimensions that can be used to divide the target population into clusters each having similar body dimensions.
Different body positions have different morphological characteristics, and existing classification criteria are often rough and do not fit well to a particular body type. In addition, the existing body type classification is mainly realized by a classical statistical method.
Because these linguistic terms reflect the designer's traditional way of expressing imprecision and ambiguity, obfuscation techniques are well suited to handle this situation. On the basis, the fuzzy clustering method is used for classifying the human body shapes, and dynamic clustering results are obtained, but the results are sometimes quite different from the actual situation in the clothing design.
The rough set has strong knowledge classification capability. Therefore, a human body type classification method based on a fuzzy rough set is provided so as to accurately classify the forms of all body parts of a target population, and the classification method is sensitive to the overall form characteristics of the target population. Meanwhile, the data related to the human body shape provided by the database is also integrated into the established classification method.
Disclosure of Invention
The invention aims to provide a human body type classification method for clothing design.
The technical scheme for solving the technical problems is as follows:
a human body type classification method for clothing design comprises the following steps:
step 1, determining the size of a key body part of a human body, and measuring the size of the key body part of the human body to be evaluated, wherein the size of each part is a characteristic index value;
step 2, comparing the picture of the body type of the human body to be evaluated with picture data in a body type classification database, and performing category evaluation on the sizes of key body parts of the body type of the human body to be evaluated to obtain a preliminary evaluation category of the sizes of each key body part, wherein the preliminary evaluation categories comprise five categories, and the categories are respectively as follows: very small, medium, large, very large;
step 3, establishing a decision table according to the characteristic index values and the preliminary classification categories, and establishing a fuzzy classification model of the body according to the decision table;
and 4, classifying the body type of the human body to be evaluated according to the fuzzy classification model and the classification algorithm of the body.
Further, in step 3, the step of establishing the fuzzy classification model is:
step 3.1, let b ═ b1，b2，…，bnIs a human body set, biThe j-th characteristic index value of (2) is xijB, carrying out the following steps of; according to the formula
Figure BDA0002272606070000021
X is to beijNormalized to
Figure BDA0002272606070000022
Wherein i is 1, …, n; j is 1, …, q, min is xi1-xiqMax is xi1-xiqMaximum value of (1);
step 3.2, calculating the values of the following five endpoints:
Figure BDA0002272606070000031
step 3.3,According to Xj1-Xj5Establishing triangular or trapezoidal membership functions
Figure BDA0002272606070000032
Will be provided with
Figure BDA0002272606070000033
Expressed as fuzzy sets
Figure BDA0002272606070000034
Index value
Figure BDA0002272606070000035
Membership function of if
Figure BDA0002272606070000036
Then
Figure BDA0002272606070000037
Is blurred into
Figure BDA0002272606070000038
Respectively convert x intoi1-xiqFall into fuzzy sets
Figure BDA0002272606070000039
Figure BDA00022726060700000310
To evaluate a corresponding fuzzy set having a rating of "very small",
Figure BDA00022726060700000311
to evaluate a fuzzy set with a "small" level of correspondence,
Figure BDA00022726060700000312
to evaluate the fuzzy sets with the "medium" correspondence level,
Figure BDA00022726060700000313
to evaluate a fuzzy set with a "large" level correspondence,
Figure BDA00022726060700000314
to evaluate a corresponding fuzzy set with a "very large" level.
Further, in step 4, the body shape classification algorithm includes the steps of:
step 4.1, if xi1-xiqAre all divided into the same fuzzy set
Figure BDA00022726060700000315
Then directly will
Figure BDA00022726060700000316
The corresponding evaluation level is used as a classification category; if xi1-xiqNot divided into the same fuzzy set
Figure BDA00022726060700000317
The next step is carried out;
step 4.2, a discrete decision table is established for each human body to be evaluated, the decision table takes key body part characteristic index values of the human body to be evaluated as conditional attributes, takes the preliminary evaluation category of the key body part size of the human body to be evaluated as decision attributes, and calculates Importance _ degree (a) according to the decision table, wherein the Importance calculation formula is as follows:
Figure BDA00022726060700000318
wherein U is the total data set in each human body decision table to be evaluated, C is the condition attribute set, D ═ D is the decision attribute set, posC(D) Is the positive domain of C of D, "| · |" represents the number of elements in the set;
determining xi1-xiqComparing the feature index values pairwise, if the importance degrees of the two feature index values are different and the difference value of the importance degrees of the two feature index values is larger than a threshold th, selecting the feature index value with higher importance degree, and finally selecting the fuzzy set corresponding to the selected feature value
Figure BDA00022726060700000319
The corresponding evaluation level is the classification category; if the importance degrees of two or more feature index values are the same or the difference value of the importance degrees is less than or equal to the threshold th, comparing the similarity of each feature index value with the mean value of all feature index values, selecting one feature index value with the highest similarity, and finally selecting the fuzzy set corresponding to the feature value
Figure BDA0002272606070000041
The corresponding evaluation level is the classification category; the calculation formula of the similarity is as follows:
Sim(I，I(j))＝exp(-||I-I(j)||)
wherein I ═ x1，x2，...xt) The indicator vector is a specific body type indicator vector, i (j) ═ (x (j),. and x (j)) are mean vectors of j-th indicator level, x (j) is an average value of all feature indicator values at the j-th indicator level, and "| · |" represents the euclidean distance between two vectors.
Further, the key body part sizes include a waist configuration, a hip configuration, an abdomen configuration, a leg length, a thigh configuration, and a calf configuration.
Further, the body type classification database stores the evaluated and classified human body type picture data.
Further, in the step 2, the picture of the body type of the human body to be evaluated is compared with the picture data in the body type classification database, and the category corresponding to the picture with the closest body type is the preliminary evaluation category.
The invention has the beneficial effects that: the invention provides a human body type classification method, which classifies the body types of target people into different categories. The invention establishes a decision table for human body type classification based on the obtained fuzzy data and database comparison data. The invention can classify the body types of human bodies and is used as reference for designing clothes.
Drawings
FIG. 1 is a schematic overview of the process of the present invention;
FIG. 2 is a schematic diagram of a fuzzy set building method when building a fuzzy classification model according to the present invention.
Detailed Description
The principles and features of this invention are described below in conjunction with the following drawings, which are set forth by way of illustration only and are not intended to limit the scope of the invention.
As shown in fig. 1, the method of the present invention comprises the steps of:
a human body type classification method for clothing design comprises the following steps:
step 1, determining the size of a key body part of a human body, and measuring the size of the key body part of the human body to be evaluated, wherein the size of each part is a characteristic index value;
step 2, comparing the picture of the body type of the human body to be evaluated with picture data in a body type classification database, and performing category evaluation on the sizes of key body parts of the body type of the human body to be evaluated to obtain a preliminary evaluation category of the sizes of each key body part, wherein the preliminary evaluation categories comprise five categories, and the categories are respectively as follows: very small, medium, large, very large;
step 3, establishing a decision table according to the characteristic index values and the preliminary classification categories, and establishing a fuzzy classification model of the body according to the decision table;
and 4, classifying the body type of the human body to be evaluated according to the fuzzy classification model and the classification algorithm of the body.
As an embodiment, in the step 3, the step of establishing the fuzzy classification model includes:
step 3.1, let b ═ b1，b2，…，bnIs a human body set, biThe j-th characteristic index value of (2) is xijB, carrying out the following steps of; according to the formula
Figure BDA0002272606070000051
X is to beijNormalized to
Figure BDA0002272606070000052
Wherein i is 1, …, n; j is 1, …, q, min is xi1-xiqMax is xi1-xiqMaximum value of (1);
step 3.2, calculating the values of the following five endpoints:
Figure BDA0002272606070000053
step 3.3, as shown in FIG. 2, according to Xj1-Xj5Establishing triangular or trapezoidal membership functions
Figure BDA0002272606070000054
Will be provided with
Figure BDA0002272606070000055
Expressed as fuzzy sets
Figure BDA0002272606070000056
Index value
Figure BDA0002272606070000057
Membership function of if
Figure BDA0002272606070000058
Then
Figure BDA0002272606070000059
Is blurred into
Figure BDA00022726060700000510
Respectively convert x intoi1-xiqFall into fuzzy sets
Figure BDA00022726060700000511
Figure BDA00022726060700000512
To evaluate a corresponding fuzzy set having a rating of "very small",
Figure BDA00022726060700000513
to evaluate a fuzzy set with a "small" level of correspondence,
Figure BDA00022726060700000514
to evaluate the fuzzy sets with the "medium" correspondence level,
Figure BDA00022726060700000515
to evaluate a fuzzy set with a "large" level correspondence,
Figure BDA00022726060700000516
to evaluate a corresponding fuzzy set with a "very large" level.
In step 4, as an embodiment, the body shape classification algorithm includes the steps of:
step 4.1, if xi1-xiqAre all divided into the same fuzzy set
Figure BDA0002272606070000061
Then directly will
Figure BDA0002272606070000062
The corresponding evaluation level is used as a classification category; if xi1-xiqNot divided into the same fuzzy set
Figure BDA0002272606070000063
The next step is carried out;
step 4.2, a discrete decision table is established for each human body to be evaluated, the decision table takes key body part characteristic index values of the human body to be evaluated as conditional attributes, takes the preliminary evaluation category of the key body part size of the human body to be evaluated as decision attributes, and calculates Importance _ degree (a) according to the decision table, wherein the Importance calculation formula is as follows:
Figure BDA0002272606070000064
wherein U is the total data set in each human body decision table to be evaluated, C is the condition attribute set, D ═ D is the decision attribute set, posC(D) Is the positive domain of C of D, "| · |" represents the number of elements in the set;
determining xi1-xiqOf importance ofComparing the multiple characteristic index values pairwise, if the importance degrees of the two characteristic index values are different and the difference value of the importance degrees of the two characteristic index values is larger than a threshold th, selecting the characteristic index value with higher importance degree, and finally selecting the fuzzy set corresponding to the characteristic value
Figure BDA0002272606070000065
The corresponding evaluation level is the classification category; if the importance degrees of two or more feature index values are the same or the difference value of the importance degrees is less than or equal to a threshold th, the threshold th is set according to experience, the similarity of each feature index value and the mean of all feature index values is compared, one feature index value with the highest similarity is selected, and finally the fuzzy set corresponding to the selected feature value is selected
Figure BDA0002272606070000066
The corresponding evaluation level is the classification category; the calculation formula of the similarity is as follows:
Sim(I，I(j))＝exp(-||I-I(j)||)
wherein I ═ x1，x2，...xt) The indicator vector is a specific body type indicator vector, i (j) ═ (x (j),. and x (j)) are mean vectors of j-th indicator level, x (j) is an average value of all feature indicator values at the j-th indicator level, and "| · |" represents the euclidean distance between two vectors.
As an embodiment, the key body part dimensions include waist morphology, hip morphology, abdomen morphology, leg length, thigh morphology, and calf morphology.
In one embodiment, the body type classification database stores the evaluated and classified human body type picture data.
As an implementation manner, in the step 2, the picture of the body type of the human body to be evaluated is compared with the picture data in the body type classification database, and the category corresponding to the picture with the closest body type is the preliminary evaluation category.
This embodiment exemplifies classification of the lower body types:
1. the key parts of the lower body are determined. The lower body shape is typically described by fashion designers using waist morphology (WS), hip morphology (HS), abdomen morphology (AS), Leg Length (LL), thigh morphology (TS), and calf morphology (CS).
2. Obtaining dimensions related to the lower body includes vertical dimensions such as: height (S), Waist Height (WH), Knee Height (KH), Thigh Length (TL), etc., and horizontal dimensions, such as: waist (W), hip (H), abdomen (A), thigh (T), shank (C), knee (K) [15 ].
3. And selecting important dimensions as measurement vectors. Knee Height (KH), Thigh Length (TL), and knee circumference (K) are more relevant to the construction of the garment pattern, while there is less concern about the design and size of the garment. In this case, only eight measurements relating to the pant design are considered as critical dimensions, including height, waist, hip, abdomen, thigh, calf, etc., which constitute the measurement vector for the particular person in the study.
4. And determining the body type index.
(1) Waist type index: WS is W/S
(2) Hip type index: HS 1-H-W, HS 2-H/W, HS 3-H/S
(3) Abdominal type index: AS 1-a-W, AS 2-a/W, AS 3-a/S
(4) Leg length index: LL1 CH/S, LL2 WH/S
(5) Thigh type index: TS 1T/W, TS 2T/H, TS 3T/S
(6) Calf type index: CS1 ═ C/W, CS2 ═ C/H, CS3 ═ C/S
5. Comparing the body shape picture with the pictures in the database, and dividing each body part into five primary evaluation categories: very Small (VS); small (S); medium (M); large (L); very Large (VL).
6. And establishing a fuzzy classification model of the lower half body. The most suitable classification index is determined in order to model the classification of the body shape.
7. And establishing a discrete decision table. And establishing a discrete decision table for each lower body by taking the relevant characteristic indexes (measured values) as conditional attributes and taking the preliminary evaluation category as a decision attribute. While all feature index values should be obscured by an appropriate set of ambiguities.
1) And classifying the lower body type according to the fuzzification of the characteristic indexes and a classification algorithm to obtain a classification result.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.

Claims (6)

1.一种用于服装设计的人体体型分类方法，其特征在于，包括以下步骤：1. a human body shape classification method for clothing design, is characterized in that, comprises the following steps: 步骤1、确定人体关键身体部位尺寸，对待评估人体关键身体部位尺寸进行测量，每个部位的尺寸为一个特征指标值；Step 1. Determine the size of the key body parts of the human body, measure the size of the key body parts of the human body to be evaluated, and the size of each part is a characteristic index value; 步骤2、将待评估人体体型图片与体型分类数据库内的图片数据进行对比，对待评估人体体型的关键身体部位尺寸进行类别评估，得到每个关键身体部位尺寸的初步评估类别，所述初步评估类别包括五种，分别为：非常小，小，中等，大，非常大；Step 2. Compare the image of the body shape to be evaluated with the image data in the body shape classification database, perform category evaluation on the size of the key body parts of the human body to be evaluated, and obtain a preliminary evaluation category of the size of each key body part, the preliminary evaluation category Including five kinds, namely: very small, small, medium, large, very large; 步骤3、根据特征指标值和初步分类类别建立决策表，根据决策表建立身体的模糊分类模型；Step 3, establishing a decision table according to the characteristic index value and the preliminary classification category, and establishing a fuzzy classification model of the body according to the decision table; 步骤4、根据身体的模糊分类模型以及分类算法对待评估人体体型进行分类。Step 4. Classify the body shape of the human body to be evaluated according to the fuzzy classification model of the body and the classification algorithm. 2.根据权利要求1所述的用于服装设计的人体体型分类方法，其特征在于，所述步骤3中，建立模糊分类模型的步骤为：2. the human body shape classification method for clothing design according to claim 1, is characterized in that, in described step 3, the step of establishing fuzzy classification model is: 步骤3.1、令b＝{b1，b2，…，bn}为人体集合，bi的第j个特征指标值为xij，；根据公式
Figure FDA0002272606060000011
将xij归一化为
Figure FDA0002272606060000012
其中i＝1，…，n；j＝1，…，q，min为xi1-xiq中的最小值，max为xi1-xiq中的最大值；Step 3.1. Let b= { b 1 , b 2 , .
Figure FDA0002272606060000011
Normalize x ij to
Figure FDA0002272606060000012
where i=1,...,n; j=1,...,q, min is the minimum value in x i1 -x iq , max is the maximum value in x i1 -x iq ; 步骤3.2、计算以下五个端点的值:Step 3.2. Calculate the values of the following five endpoints:
Figure FDA0002272606060000013
Xj4＝Xj3+Xj52,Xj5＝max1≤i≤nxij；
Figure FDA0002272606060000013
X j4 =Xj3+Xj52, Xj5=max1≤i≤nxij; 步骤3.3、根据Xj1-Xj5建立三角形或梯形隶属函数
Figure FDA0002272606060000014
将
Figure FDA0002272606060000015
表示为模糊集
Figure FDA0002272606060000016
索引值
Figure FDA0002272606060000017
的隶属函数，如果
Figure FDA0002272606060000018
则
Figure FDA0002272606060000019
被模糊化为
Figure FDA00022726060600000110
分别将xi1-xiq归入模糊集
Figure FDA0002272606060000021
Step 3.3. Establish a triangular or trapezoidal membership function according to X j1 -X j5
Figure FDA0002272606060000014
Will
Figure FDA0002272606060000015
represented as a fuzzy set
Figure FDA0002272606060000016
index value
Figure FDA0002272606060000017
membership function, if
Figure FDA0002272606060000018
but
Figure FDA0002272606060000019
blurred to
Figure FDA00022726060600000110
Respectively classify x i1 -x iq into fuzzy sets
Figure FDA0002272606060000021
Figure FDA0002272606060000022
为评估级别为“非常小”对应的模糊集，
Figure FDA0002272606060000023
为评估级别为“小”对应的模糊集，
Figure FDA0002272606060000024
为评估级别为“中”对应的模糊集，
Figure FDA0002272606060000025
为评估级别为“大”对应的模糊集，
Figure FDA0002272606060000026
为评估级别为“非常大”对应的模糊集。
Figure FDA0002272606060000022
is the fuzzy set corresponding to the evaluation level "very small",
Figure FDA0002272606060000023
is the fuzzy set corresponding to the evaluation level "small",
Figure FDA0002272606060000024
is the fuzzy set corresponding to the evaluation level "Medium",
Figure FDA0002272606060000025
is the fuzzy set corresponding to the evaluation level "large",
Figure FDA0002272606060000026
Fuzzy set corresponding to the evaluation level "Very Large". 3.根据权利要求1所述的用于服装设计的人体体型分类方法，其特征在于，所述步骤4中，身体形状分类算法的步骤为：3. the human body shape classification method that is used for clothing design according to claim 1, is characterized in that, in described step 4, the step of body shape classification algorithm is: 步骤4.1、若xi1-xiq均被分入了同一个模糊集
Figure FDA0002272606060000027
则直接将
Figure FDA0002272606060000028
对应的评估级别作为分类类别；若xi1-xiq未被分入同一个模糊集
Figure FDA0002272606060000029
则转入下一步；Step 4.1. If x i1 -x iq are all classified into the same fuzzy set
Figure FDA0002272606060000027
directly
Figure FDA0002272606060000028
The corresponding evaluation level is used as the classification category; if x i1 -x iq are not classified into the same fuzzy set
Figure FDA0002272606060000029
then go to the next step; 步骤4.2、为每个待评估人体建立一个离散的决策表，决策表以待评估人体的关键身体部位特征指标值为条件属性，以待评估人体关键身体部位尺寸的初步评估类别为决策属性，根据决策表计算重要度Importance_degree(a)，重要度计算公式为：Step 4.2. Establish a discrete decision table for each human body to be evaluated. The decision table takes the feature index of the key body parts of the human body to be evaluated as the condition attribute, and takes the preliminary evaluation category of the size of the key body parts of the human body to be evaluated as the decision attribute. The decision table calculates Importance_degree(a), and the formula for calculating the importance is:
Figure FDA00022726060600000210
Figure FDA00022726060600000210
 其中U是每个待评估人体决策表内总的数据集合，C是条件属性集，D＝{d}是决策属性集，posC(D)是D的C的正域，“|·|”表示集合中元素的个数；where U is the total data set in the decision table of each human being to be evaluated, C is the condition attribute set, D={d} is the decision attribute set, pos C (D) is the positive field of C of D, "|·|" Represents the number of elements in the collection; 确定xi1-xiq的重要度，将多个特征指标值进行两两对比，若两个特征指标值的重要程度不相同且二者的重要度差值大于阈值th,则选择重要程度较高的特征指标值，最终选择的特征值对应的模糊集
Figure FDA00022726060600000211
对应的评估级别即为分类类别；若两个或多个特征指标值的重要程度一样或重要度差值小于或等于阈值th,则比较每个特征指标值与所有特征指标值均值的相似度，选择相似度最高的一个特征指标值，最终选择的特征值对应的模糊集
Figure FDA00022726060600000212
对应的评估级别即为分类类别；所述相似度的计算公式为：Determine the importance of x i1 -x iq , and compare multiple feature index values in pairs. If the importance levels of the two feature index values are not the same and the difference in importance between the two is greater than the threshold th, select the higher level of importance. The characteristic index value of , the fuzzy set corresponding to the final selected characteristic value
Figure FDA00022726060600000211
The corresponding evaluation level is the classification category; if the importance of two or more feature index values is the same or the difference in importance is less than or equal to the threshold th, then compare the similarity between each feature index value and the mean value of all feature index values, Select a feature index value with the highest similarity, and finally select the fuzzy set corresponding to the feature value
Figure FDA00022726060600000212
The corresponding evaluation level is the classification category; the calculation formula of the similarity is: Sim(I，I(j))＝exp(-||I-I(j)||)Sim(I, I(j))=exp(-||I-I(j)||) 其中I＝(x1，x2，...xt)为一个特定的体型指标向量，I(j)＝(x(j)，x(j)，...，x(j))为第j个指标水平的均值向量，x(j)是第j个指标水平上所有特征指标值的平均值，“||·||”表示两个向量的欧式距离。where I=(x 1 , x 2 ,...x t ) is a specific body shape indicator vector, and I(j)=(x(j), x(j),...,x(j)) is The mean vector of the jth indicator level, x(j) is the average value of all feature indicator values on the jth indicator level, and "||·||" represents the Euclidean distance of the two vectors. 4.根据权利要求1所述的用于服装设计的人体体型分类方法，其特征在于，所述关键身体部位尺寸包括腰部形态、臀部形态、腹部形态、腿长、大腿形态和小腿形态。4. The human body shape classification method for clothing design according to claim 1, wherein the key body part dimensions include waist shape, buttock shape, abdomen shape, leg length, thigh shape and calf shape. 5.根据权利要求1所述的用于服装设计的人体体型分类方法，其特征在于，所述体型分类数据库内存储有已评估分类的人体体型图片数据。5 . The human body shape classification method for clothing design according to claim 1 , wherein the body shape classification database stores the evaluated and classified human body shape picture data. 6 . 6.根据权利要求1所述的用于服装设计的人体体型分类方法，其特征在于，所述步骤2中，将待评估人体体型图片与体型分类数据库内的图片数据进行对比，体型最接近的一张图片对应的类别即为初步评估类别。6. the human body shape classification method for clothing design according to claim 1, is characterized in that, in described step 2, the body shape picture to be evaluated is compared with the picture data in the body shape classification database, the body shape closest to is compared. The category corresponding to a picture is the preliminary evaluation category.
CN201911110639.4A 2019-11-14 2019-11-14 Human body type classification method for clothing design Active CN111126432B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201911110639.4A CN111126432B (en) 2019-11-14 2019-11-14 Human body type classification method for clothing design

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201911110639.4A CN111126432B (en) 2019-11-14 2019-11-14 Human body type classification method for clothing design

Publications (2)

Publication Number Publication Date
CN111126432A true CN111126432A (en) 2020-05-08
CN111126432B CN111126432B (en) 2023-07-07

Family

ID=70495591

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201911110639.4A Active CN111126432B (en) 2019-11-14 2019-11-14 Human body type classification method for clothing design

Country Status (1)

Country Link
CN (1) CN111126432B (en)

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20060167350A1 (en) * 2005-01-27 2006-07-27 Monfre Stephen L Multi-tier method of developing localized calibration models for non-invasive blood analyte prediction
US20070032898A1 (en) * 2001-05-11 2007-02-08 Wang Kenneth K Method and apparatus for identifying vitual body profiles
US20110052005A1 (en) * 2009-08-28 2011-03-03 Allen Joseph Selner Designation of a Characteristic of a Physical Capability by Motion Analysis, Systems and Methods
CN105894026A (en) * 2016-03-31 2016-08-24 东华大学 Figure classifying method based on fuzzy theory
CN105956382A (en) * 2016-04-26 2016-09-21 北京工商大学 Traditional Chinese medicine constitution optimized classification method based on improved CART decision-making tree and fuzzy naive Bayes combined model
CN107146148A (en) * 2017-05-26 2017-09-08 重庆慧高科技有限公司 Operation system based on human outside data acquisition
CN107203753A (en) * 2017-05-25 2017-09-26 西安工业大学 A kind of action identification method based on fuzzy neural network and graph model reasoning
CN108920822A (en) * 2018-06-30 2018-11-30 赵志泓 A kind of template rapid generation
CN109255363A (en) * 2018-07-11 2019-01-22 齐鲁工业大学 A kind of fuzzy k nearest neighbor classification method and system based on weighted chi-square distance metric

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20070032898A1 (en) * 2001-05-11 2007-02-08 Wang Kenneth K Method and apparatus for identifying vitual body profiles
US20060167350A1 (en) * 2005-01-27 2006-07-27 Monfre Stephen L Multi-tier method of developing localized calibration models for non-invasive blood analyte prediction
US20110052005A1 (en) * 2009-08-28 2011-03-03 Allen Joseph Selner Designation of a Characteristic of a Physical Capability by Motion Analysis, Systems and Methods
CN105894026A (en) * 2016-03-31 2016-08-24 东华大学 Figure classifying method based on fuzzy theory
CN105956382A (en) * 2016-04-26 2016-09-21 北京工商大学 Traditional Chinese medicine constitution optimized classification method based on improved CART decision-making tree and fuzzy naive Bayes combined model
CN107203753A (en) * 2017-05-25 2017-09-26 西安工业大学 A kind of action identification method based on fuzzy neural network and graph model reasoning
CN107146148A (en) * 2017-05-26 2017-09-08 重庆慧高科技有限公司 Operation system based on human outside data acquisition
CN108920822A (en) * 2018-06-30 2018-11-30 赵志泓 A kind of template rapid generation
CN109255363A (en) * 2018-07-11 2019-01-22 齐鲁工业大学 A kind of fuzzy k nearest neighbor classification method and system based on weighted chi-square distance metric

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
何亚群,胡寿松: "一种基于粗糙-模糊集集成模型的决策分析方法", no. 03, pages 76 - 79 *
杜瑞卿: "粗糙集模糊聚类分析法在昆虫分类研究中的应用", pages 106 - 111 *
陈敏之: "女体体型识别专家系统的设计与实现", pages 142 - 144 *

Also Published As

Publication number Publication date
CN111126432B (en) 2023-07-07

Similar Documents

Publication Publication Date Title
CN101419671B (en) Face gender identification method based on fuzzy support vector machine
CN103886026B (en) Clothing matching method based on personal feature
Chung et al. The development of sizing systems for Taiwanese elementary-and high-school students
Liu et al. Fuzzy classification of young women's lower body based on anthropometric measurement
CN109919177B (en) Feature selection method based on hierarchical deep network
CN109101938B (en) Multi-label age estimation method based on convolutional neural network
US20110142335A1 (en) Image Comparison System and Method
CN106055893A (en) Clothes matching scheme generation method based on fashion template database and automatic matching
CN104992142A (en) Pedestrian recognition method based on combination of depth learning and property learning
CN105956382A (en) Traditional Chinese medicine constitution optimized classification method based on improved CART decision-making tree and fuzzy naive Bayes combined model
CN108960043A (en) A kind of personage's family relationship construction method for electron album management
CN108460407A (en) A kind of pedestrian's attribute fining recognition methods based on deep learning
CN109145947B (en) Fashion women's dress image fine-grained classification method based on part detection and visual features
CN107885778A (en) A kind of personalized recommendation method based on dynamic point of proximity spectral clustering
CN108897805A (en) A kind of patent text automatic classification method
CN109635140A (en) A kind of image search method clustered based on deep learning and density peaks
Chi et al. A study on segmentation and refinement of key human body parts by integrating manual measurements
CN104899280A (en) Fuzzy-correlated asynchronous image retrieval method based on color histogram and NSCT (Non-Subsampled Contourlet Transform)
Xue et al. A method of the product form design and color design of train seats based on GRA-Fuzzy theory
CN111126432A (en) A Human Body Type Classification Method for Clothing Design
Feifei et al. Multi-core SVM optimized visual word package model for garment style classification
Yao et al. Classification of female body shape based on two-dimensional image and computer vision technology
CN110222564A (en) A method of sex character is identified based on three-dimensional data
Jin et al. Developing an intelligent model for the construction a hip shape recognition system based on 3D body measurement
Kim et al. An effective research method to predict human body type using an artificial neural network and a discriminant analysis

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant