JPH08285838A - Method and system for classifying fabric - Google Patents

Abstract

PURPOSE: To obtain method and system for classifying fabric in which a new fabric can be classified into an existing group uniformly with high reliability. CONSTITUTION: The fabric classifying system comprises means 3 for storing the standard data of objective fabric and standard fabric, and means 4 for calculating a feature data, represented by the deterministic values related to each feature element of fabric, from the standard data 3. The fabric classifying system further comprises means 5 for storing the feature data of objective fabric and standard fabric calculated by the feature data calculating means 4, and means 6 for retrieving a standard fabric resembling the objective fabric most from the feature data stored in the feature data memory means 5.

Description

Detailed Description of the Invention

[0001]

FIELD OF THE INVENTION The present invention relates to a method and apparatus for classifying textiles based on textile specification data.

[0002]

2. Description of the Related Art When formulating sales plans and product strategies in the manufacturing industry, it is indispensable to classify all products based on certain rules and to grasp the sales tendency for each classification. Especially in the textile manufacturing industry, the necessity of this classification is high. By the way, the conventional classification of textiles relies on human judgment. That is, in the past, experts who had specialized knowledge and know-how related to the design and manufacturing of textiles,
When registering textile standard data such as organization, density, processing,
The method is to determine which group of existing textiles the textile should belong to and to register the code of that group in a ledger, computer or the like.

[0003]

However, when judging by personal knowledge, different classification codes are assigned to textiles having exactly the same standard data depending on the person who judges, and even one expert sometimes uses the same standard code. It even assigned different classification codes to the textiles. In this way, since the results of the textile classification lacked homogeneity and reliability, it was not always possible to provide effective information when formulating sales plans and product strategies.

The present invention solves such a problem by assigning a new fabric to a group of existing fabrics,
An object of the present invention is to provide a method and a device for classifying textiles, which can classify the materials uniformly and with high reliability.

[0005]

The invention described in claim 1 of the present invention for solving the above-mentioned problems is a method for classifying a woven fabric by allocating the woven fabric to a standard woven fabric based on the standard data of the woven fabric. For the classification target fabric and the standard fabric, the characteristic data represented by the confidence value for each characteristic element of the fabric is calculated from the standard data, and then the calculated target data of the classification target fabric and the standard fabric The standard woven fabric that is most similar to the woven fabric is searched, and the invention according to claim 2 is a method of allocating the woven fabric to the standard woven fabric based on the standard data of the woven fabric and classifying the woven fabric. Therefore, first, with respect to the woven fabric to be classified and the standard woven fabric, the characteristic data represented by the confidence value regarding each characteristic element of the woven fabric is calculated from the standard data, and then before the calculation. Characteristic data of a classification target fabric and a standard fabric are arranged in an n-dimensional space, and a standard fabric having a characteristic data value closest to the feature data value of the classification target fabric in the space is selected as a similar fabric. It is characterized by.

According to a third aspect of the present invention, there is provided an apparatus for classifying a woven fabric into a standard woven fabric based on the woven fabric standard data, and storing the woven fabric to be classified and the standard data of the standard woven fabric. Storage means, characteristic data calculation means for calculating characteristic data represented by certainty values for each characteristic element of the fabric from the standard data, and characteristic data of the classification target fabric and the standard fabric calculated by the characteristic data calculation means And a similar fabric retrieval unit for retrieving the standard fabric most similar to the classification target fabric from the feature data stored in the feature data storage unit. According to a fourth aspect of the invention, the similar fabric searching means in the third aspect of the invention uses the n-dimensional empty space for the feature data of the classification target fabric and the standard fabric. Disposed within, is that selects standard fabric with the characteristic data value at the closest distance between characteristic data values of the classification target fabric within the space as a similar fabric.

The standard data of the woven fabric is used to define the manufacturing process of the woven fabric, such as the raw yarns, the woven fabric structure, the density of the warp yarns and the weft yarns (the number of yarns / inch), the processing method of the woven fabric, etc. The data required for Further, the standard woven fabric refers to a core representative woven fabric in which an existing woven fabric is divided into groups by a person skilled in the art such as a fabric designer according to an empirical rule. In addition, the characteristic data of the woven fabric refers to a characteristic element indicating a characteristic of the woven fabric with a certainty value, and the characteristic elements are “plain weave”, “fine count”, “clear”, “twill”, “ An element (item) that characterizes a woven fabric, such as “high density” or “milled”, and the certainty value is a numerical value of 0 or more and 1 or less that the characteristic element indicating the woven fabric characteristic is appropriate as the woven fabric characteristic. What is expressed.

[0008]

Next, the operation of the present invention will be described. First, the invention according to claims 1 and 3 will be described. According to this invention, the characteristic data represented by the certainty value in each element of the standard data is calculated for the classification target fabric and the standard fabric. As a result, it is possible to express the characteristics of the fabric by numerical values. Then, the standard fabric most similar to the classification target fabric is searched from the calculated characteristic data of the classification target fabric and the standard fabric. At this time, since the features of the fabric are represented by numerical values as described above, it is extremely easy to search for a similar standard fabric.

Further, according to the present invention, when it is desired to set a new standard fabric (classification group), the characteristic data of the standard fabric can be calculated by the characteristic data calculating means and registered in the characteristic data storing means. Further, when it is desired to change the standard fabric already registered in the feature data storage means to another fabric, the feature data of the fabric may be calculated and stored in the feature data storage means. ， There is no need to rearrange complicated classification logic when adding, and it is possible to easily change and add.

Next, the invention according to claims 2 and 4 will be described. According to the present invention, when searching for the standard fabric most similar to the classification target fabric, the calculated characteristic data values of the classification target fabric and the standard fabric are arranged in the n-dimensional space, and the classification target in the space is classified. A standard fabric having a characteristic data value closest to the characteristic data value of the fabric is selected as a similar fabric. Thus, by expressing the fabrics as coordinates in the n-dimensional space, the degree of similarity between the fabrics and the most similar fabric can be obtained by numerical calculation.

[0011]

Embodiments of the present invention will be described below with reference to the accompanying drawings. In addition, in this example, the classification for woolen fabrics will be described, but it goes without saying that the classification target is not limited to this.

As shown in FIG. 1, the embodiment apparatus has an input means (1) including a keyboard, a processing device (2) including a memory, a hard disk, a CPU, etc., and an output means (7) including a monitor, a printer and the like. ) And. Also,
The processing device (2) comprises standard data storage means (3), characteristic data calculation means (4), characteristic data storage means (5), and similar fabric retrieval means (6). The details of each unit will be described below, but since the input unit (1) and the output unit (7) have known configurations, detailed description thereof will be omitted.

(Standard Data Storage Means) The standard data storage means (3) is composed of a memory, a hard disk, etc., and the standard data of the fabric inputted from the input means (1) is used for identifying the fabric. This is a means for storing the product number as a key. It should be noted that the standard data are various data necessary for manufacturing the woven fabric such as the types of the warp and the weft, the woven structure, the woven density, and the finishing method as described above. An example of the standard data stored in the standard data storage means (3) is shown in FIG.
The woven fabric of product number 100 shown in the figure uses the yarn of KB as the warp and the weft of the brand, the number of the indicator is 2/72, the weave is plain weave, the warp density of the weave density is 65.5 yarns / inch, the weft density 54.5 pieces / inch, gas baking for finishing, washing,
The steps of boiling, drying, shaving, pressing, and kettle steaming are carried out in sequence. Further, the woven fabric of the product number 200 uses BE for the warp and weft as the brand, 2/48 for the display count, the twill weave has a warp density of 56.0 /
Inch, weft density is 51.0 yarns / inch, and finishing processes include washing, crimping, drying, raising, shearing, pressing, and steaming.

(Characteristic Data Calculation Means) The characteristic data calculation means (4) calculates characteristic data in which the characteristic and properties of the fabric are converted into numerical data from the standard data of the fabric stored in the standard data storage means (3). Is. Here, the characteristic data will be described. When a person skilled in the art expresses a woven fabric, for example, the expressions such as “woven fabric A is a plain weave fine count fabric” and “woven fabric B is a twill weave and a high density milled fabric” are used. Characteristic elements such as "fine count", "clear", "twill", "high density" and "milled" are used. The characteristic data is a numerical value indicating the suitability of the woven fabric to these characteristic elements, that is, whether or not the woven fabric is applicable to the characteristic elements. Here, this matching degree is referred to as a "confidence value". If the "characteristic element" is extremely suitable for expressing a fabric, its confidence value is 1.0, if it is completely inappropriate, 0.0, and if it is to some extent, depending on its degree. A proper value of 0.0 to 1.0 is set. Thus, for example, if the certainty values of "fine count" for a certain woven fabric A and B are 0.5 and 0.8, respectively, it means that the woven fabric B is a fine count than the woven fabric A.

The method of calculating the certainty value will be described below. Since the calculation method differs depending on each "characteristic element", description will be made separately for each type.

(A) First, "fine count", "thick count",
Items that can be numerically calculated using standard data such as “high density” and “low density” will be described. For example, FIGS.
As shown in, the correlations shown in the figure can be obtained from the empirical rule of the fabric designer, for the "fine count" and the average count of the weaving yarn, and for the "high density", the count density ratio. In this example, a linear relationship is exemplified, but the present invention is not limited to this, and a non-linear relationship is naturally acceptable.

According to FIG. 3, if the average yarn count of the weaving yarn is 38 or above, the certainty value of "fine count" is 1.0,
If it is 28th or less, the certainty value is 0.0, and between 28th and 38th, if Ya is the certainty value and Xa is the average number, the certainty value Ya is the following equation: Ya = 0.1Xa It can be expressed as -2.8. According to this, since the average count of the woven fabric of the product number 100 shown in FIG. 2 is 36, the “fine count” is obtained from the above formula.
The belief value Ya that is: Ya = 0.1 × 36−2.8 = 0.8

According to FIG. 4, the count density ratio is 12.
If it is 0 or more, the belief value of “high density” is 1.0, and if it is 8.0 or less, the belief value is 0.0.
In the range from 1 to 12.0, where Yb is the certainty value and Xb is the count density ratio, the certainty value Yb can be expressed by the following equation: Yb = 0.25Xb-2.0. The count density ratio Xb can be expressed by the following equation. Count density ratio Xb = (warp density + weft density) / {2 × (average count) ^1/2 } According to this, the woven fabric of product number 100 shown in FIG. 2 has a warp density of 65.5 yarns / inch, weft Since the density is 54.5 lines / inch, the count density ratio Xb is Xb = (65.5 + 54.5) / {2 × (36) ^1/2 }.
= 10.0, and the certainty value Yb is Yb = 0.25 × 10.0−2.0 = 0.5.

(B) Next, characteristic elements such as "plain weave", "twill weave", "clear", "milled", etc. for which a certainty value cannot be simply calculated based on a mathematical formula based on standard data will be described. These characteristic elements are not related to standard data represented by numerical values such as count and density, but are related to standard data represented by figures and symbols such as organization charts and machining process series. Logic is required to calculate the certainty value from the data. Confidence values of these characteristic elements are calculated as follows based on the criteria obtained by the empirical rule of the fabric designer.

(B-1) First, the woven structure will be described. In general, the woven structure is represented by a graphic in which white and black squares are arranged as shown in the structure column of FIG. Here, each square represents a point where the warp and the weft intersect (hereinafter, this point is referred to as an "intersection"). When the square is black, the warp appears on the fabric surface, and when the white, the weft appears on the fabric surface. . And
The state in which the warp and the weft in the weave structure are crossed is expressed using the word up and down (that is, the warp is above the weft at the intersection, or the warp is below the weft). In the following, the basic structures “plain weave”, “twill weave”, and “satin weave” will be sequentially illustrated.

(A) Plain weave The "basic weave", which is the most basic weave of a fabric, is "in the organization chart, ups and downs of intersections are not continuous up and down and left and right" (this is defined as "flat weave"). ). Here, continuous ups and downs mean that the ups and downs of an intersection are the same as the ups and downs of an intersection that is adjacent in any one of the up, down, left, and right directions. Therefore, the certainty value is calculated in consideration of such characteristics. That is, it is expressed by the ratio of the intersections satisfying the above definition among all the intersections in the minimum range of the woven fabric (the minimum unit of the structure repeating in the vertical and horizontal directions). Here, the certainty value Yc can be represented by the following equation: Yc = (number of intersections satisfying the definition of "plain weave") / (number of all intersections).

This will be described more specifically with reference to FIG. In the figure, at the intersection (A), warp yarns appear on the woven fabric surface, and at intersections adjacent to each other vertically and horizontally, weft yarns appear on the woven fabric surface. Therefore, the intersection (A) satisfies the definition of "plain weave". On the other hand, at the intersection (B), ups and downs are not continuous in the vertical direction, but ups and downs are continuous on the right side. Therefore, the intersection (B) does not satisfy the above definition of "plain weave". Further, the intersection (C) is located at the lower right corner in the organization chart, but as shown in the above, the organization chart is the smallest unit that is repeated vertically and horizontally, so that the intersection (C) The intersection next to the right is the intersection (D) at the lower left corner of the organizational chart, and the next adjacent intersection is the intersection (E) at the upper right corner of the organizational chart. Therefore, the intersection (C) is the "plain weave" above.
Satisfies the definition of. As described above, it is determined whether or not all the intersections satisfy the definition of "plain weave". In the case of the example in FIG. 5, out of the total 64 intersections, 24 intersections satisfy the definition of “plain weave”, and therefore the certainty value that the fabric is “plain weave” is 24/64 = 0.375.

Further, regarding the woven fabric of the product number 100 shown in FIG. 2, the certainty value Yc that this woven fabric is “plain weave” is Yc = (36/36) = 1.0. Similarly, regarding the woven fabric of product number 200, the certainty value Yc that this woven fabric is “plain weave” is Yc = (4/40) = 0.1. Further, regarding the twill fabric shown in FIG. 6, the certainty value Yc that this fabric is “plain weave” is Yc = 32/64 = 0.5. Further, regarding the satin woven fabric shown in FIG. 7, the certainty value Yc that this woven fabric is “plain weave” is Yc = (5/25) = 0.2.

(B) Twill Weave Next, the twill fabric will be described. The twill woven fabric is "When one row in the vertical direction is used as a reference in the organization chart, the arrangement of the intersections of the one row on the right side is such that the reference row is shifted one cell upward. (Strictly speaking, it is also considered that twill weaves are shifted in the downward direction, but here, for simplification of the explanation, only shifts in the upward direction are considered.)
Is defined as)). Therefore, the certainty value is calculated in consideration of such characteristics. That is, in the minimum range of the woven fabric, "the ratio of the row in the vertical direction to the row in the right direction that is offset by 1 square in the entire row in the entire row" is used as the standard, It is assumed that the certainty value Yc is subtracted to represent the certainty value Yd of "twill". Here, the reason why the belief value Yc that is "plain weave" is reduced is that when the above criteria is applied to the plain weave of product number 100 shown in FIG. This is because it is necessary to reduce the certainty value which is the "twill" of the plain weave. Here, the certainty value Yd that is “twill” can be expressed by Yd = {(the number of columns in which the adjacent column on the right is deviated by one square in the upward direction) / the total number of columns} × (1-Yc).

Looking specifically at the twill fabric of FIG. 6,
The certainty value Yc of "plain weave" is Yc = 32/64 = 0.5 The certainty value Yd of "twill weave" is Yd = (8/8) × (1-0.5) = 0.5 . Further, regarding the plain woven fabric of the product number 100 shown in FIG. 2, the certainty value Yc of “plain weave” is Yc = 36/36 = 1.0 The certainty value Yd of “twill weave” is Yd = (6 / 6) × (1-1) = 0. Further, regarding the woven fabric of the product number 200 shown in FIG. 2, the certainty value Yc that is “plain weave” is: Yc = 4/40 = 0.1 The certainty value Yd that is “twill weave” is the reference among all 10 columns. Since all except the two columns of the 8th and 10th columns from the left satisfy the condition, Yd = (8/10) × (1-1.0) = 0.72. As for the satin weave of FIG. 7, the certainty value Yc of “plain weave” is Yc = 5/25 = 0.2 The certainty value Yd of “twill weave” is Yd = (0/5) × (1
−0.2) = 0.0.

(C) Satin woven fabric Next, the satin woven fabric will be described. Fig. 7 shows a typical satin weave organization chart. As shown in the figure, the satin fabric "has few warps in one row in the vertical direction, and the intersections of warps are not continuous in the weft direction." For simplification, consider only the case where the warp is small.) (This is the definition of "satin fabric"). Therefore, the certainty value is calculated in consideration of such characteristics. That is, the certainty value Ye which is "satin weave" is set to "with respect to the row in the vertical direction within the minimum range of the fabric, the grid points of the warp floats are 25% or less, and the grid points of the warp floats are adjacent to the row. The columns that are not adjacent to the left and right columns that are not adjacent to the floating grid points occupy the entire column. " The certainty value Ye, which is "satin weave", can be expressed by Ye = (the number of columns satisfying the condition and / the total number of columns).

Looking specifically at the fabric shown in FIG. 7,
Since all five rows satisfy the above conditions and, the certainty value Ye of "satin weave" is Ye = (5/5) = 1.0 Similarly, the woven fabric of the product number 100 and the product number 20 shown in FIG.
The belief value Ye, which is the "satin weave" of both of the woven fabrics of 0, is 0.0.

(B-2) Next, the characteristic elements related to the machining series will be described by taking "clear" and "milled" as examples. (A) Clear “Clear” means the following four conditions for the machining series:
"Includes" gas-fired, "does not include" filament, "
It is characterized by satisfying “including buffalo” and “including sheared hair”. Therefore, the certainty value is calculated in consideration of such characteristics. That is, the certainty value Yf that is “clear” is represented by “a ratio of the number of conditions satisfying the above conditions to the following conditions”. Here, the certainty value Yf can be expressed by Yf = (number of applicable conditions) / (total number of conditions).

Specifically, regarding the woven fabric of product number 100 shown in FIG. 2, the certainty value Yf of "clear" is Yf = (4/4) = 1.0. Further, regarding the woven fabric of the product number 200, the certainty value Yf that is “clear” is Yf = (1/4) = 0.25.

(B) Milled "Milled" has the following three conditions for the machining series:
"Including" shrink ", not including" gas-fired ",
It is characterized by satisfying "does not include" niboshi ". Therefore, the certainty value is calculated in consideration of such characteristics. That is, the certainty value Yg which is "mild" is
It is expressed as "a ratio of the number of the conditions to which the applicable condition is provided". Here, the certainty value Yg can be represented by Yg = (the number of applicable conditions) / (the total number of conditions).

Specifically, regarding the woven fabric of the product number 100 in FIG. 2, the certainty value Yg of "milled" is Yg = (0/3) = 0.0. Further, regarding the woven fabric of the product number 200, the certainty value Yg of being “milled” is Yg = (2/3) = 0.67.

The characteristic data calculating means (4) is provided with a program for calculating each of the above confidence values, and by executing the program, each of the above confidence values is calculated. As described above, by standardizing the relationship between the characteristic element of the characteristic data and the standard data based on the knowledge and know-how of the expert, the characteristic data of the fabric can be calculated from the standard data.

(Characteristic Data Storage Means) The characteristic data storage means (5) comprises a memory or the like, and stores the characteristic data of the fabric calculated by the characteristic data calculation means (4) together with its product number. FIG. 8 shows an example of stored contents.
Shown in

(Similar Fabric Retrieval Means) The similar fabric retrieval means (6) is a means for allocating the fabric to be classified to the most similar standard fabric. There are various possible methods for searching for the most similar standard fabric, but in this example, the search method using spatial distance was adopted. In this search method, an n-dimensional space having the characteristic elements of the characteristic data as coordinate axes is set, and the fabric is represented by an n-dimensional vector determined by the characteristic data (that is, the confidence value of each characteristic element representing the characteristic). It is considered that the smaller the distance between the two fabrics is, the more similar the features of the two fabrics are to each other, the standard fabric having the smallest distance is searched, and the standard fabric is assigned to the classification target fabric.

There are many possible methods for calculating the distance between two vectors. Here, the case where the positive square root of the sum of squares of the difference between the values of the coordinate axes is taken is used. Therefore, if the distance between two vectors is D, the distance D is expressed by the following equation, D = {(Ya1 -Ya2) 2 + (Yb1 -Yb2) 2 +
(Yc1-Yc2) ² + ... (Yn1-Yn2
) ² } ^1/2 .

The fabric of product number 100 and the product number 200 shown in FIG.
Taken as an example, the distance D between the feature data of these two fabrics
^{Is, D = {(1.00-0.00) 2} + (0.00-1.0
0) ² + (0.80-0.00) ² + (0.00-0.
10) ² + (0.50-0.73) ² + (0.20-
0.00) ² + (1.00-0.00) ² + (0.0
0−0.80) ² } ^1/2 = 1.93. The similar fabric retrieval means (6) is provided with a program for calculating the distance D described above, and sequentially calculates the distance between the fabrics from the characteristic data of the classification target fabric and the standard fabric,
The standard fabric having the shortest distance is extracted, and the standard fabric name is output.

Fabric classification processing in the embodiment apparatus will be described below with reference to FIGS. The processing device (2) stores a program for executing the processing procedure shown in FIGS. First, as shown in FIG. 9, the standard data of the standard fabric and the classification target fabric is input to the processing device (2) by the input means (1) (step 101). Then, the processing device (2) stores the inputted standard data in the standard data storage means (3) (step 102). As described above, the standard data storage means (3) stores standard data with the article number of the fabric as a key.

Next, the processing device (2) calculates the characteristic data of the standard fabric according to the procedure shown in FIG. The standard woven fabric is a woven fabric which is recognized as a standard woven fabric by a person who is involved in the woven fabric industry, such as a woven fabric designer, based on experience. First, the characteristic data calculating means (4) inputs the standard data of the standard fabric stored in the standard data storing means (3) (step 20).
1), feature data is calculated from the standard data (step 202). Then, the characteristic data is stored in the characteristic data storage means (5) in association with the product number of the standard fabric (step 203).

Then, according to the procedure shown in FIG. 11, the processing device (2) classifies the target fabrics. First, the characteristic data calculation means (4) inputs the standard data of the classification target fabric from the standard data storage means (3) (step 301) and calculates the characteristic data of the classification target fabric (step 302),
The calculated characteristic data is stored in the characteristic data storage means (5) (step 303). Then, the similar fabric search means (6) inputs the feature data of the classification target fabric and the feature data of the standard fabric from the feature data storage means (5), and the classification target fabric and the standard fabric with respect to the feature data described above. Of the standard fabrics are sequentially calculated, and the standard fabric having the shortest distance is selected (step 30).
4) The product number of the standard fabric is output to the output means (7) (step 305). Receiving this, output means (7)
Displays the part number. As described above, the classification target fabric is classified into the closest standard fabric.

To set a new standard fabric, the feature data of the standard fabric is calculated by the feature data calculating means (4), and the feature data is stored in the feature data storing means (5).
You can register at. As described above, according to the fabric classification device of the embodiment, when adding or changing the standard fabric, it is not necessary to change the program in which the classification logic is described, and the addition and the change can be easily performed.

[0041]

According to the present invention, uniform textile classification can be performed by using a computer or the like without preparing the textile classification rules for all kinds of textiles to be classified into a form having logical consistency. That is, it is possible to exclude human arbitrariness in classification. Further, when it is desired to change the classification system such as addition of a new classification group, it can be easily dealt with by adding / changing the fabric whose characteristic data representing the fabric is registered in the feature storage means.

[Brief description of drawings]

FIG. 1 is a block diagram showing an overall configuration of a device according to an embodiment.

FIG. 2 is an explanatory diagram showing an example of standard data of a fabric.

FIG. 3 is an explanatory diagram showing an example of a relationship between an average count and a certainty value in a characteristic element “fine count”.

FIG. 4 is an explanatory diagram showing an example of a relationship between a count / density ratio and a certainty value in a characteristic element “high density”.

FIG. 5 is an explanatory view schematically showing a plain weave structure.

FIG. 6 is an explanatory diagram schematically showing a twill weave structure.

FIG. 7 is an explanatory view schematically showing a satin weave system structure.

FIG. 8 is an explanatory diagram showing an example of feature data of a fabric.

FIG. 9 is a flowchart showing a procedure for storing standard data of fabric.

FIG. 10 is a flowchart showing a procedure for calculating characteristic data of standard fabric.

FIG. 11 is a flow chart showing a procedure for classifying a fabric.

[Description of sign]

 DESCRIPTION OF SYMBOLS 1 input means 2 processor 3 standard data storage means 4 characteristic data calculation means 5 characteristic data storage means 6 similar fabric retrieval means 7 output means

Claims (4)

[Claims] 1. A method of classifying a woven fabric by allocating the woven fabric to a standard woven fabric based on standard data of the woven fabric, wherein the woven fabric to be classified and the standard woven fabric are expressed by a confidence value for each characteristic element of the woven fabric. A method for classifying a fabric, wherein data is calculated from the standard data, and then the standard fabric most similar to the target fabric is searched from the calculated feature data of the target fabric and the standard fabric thus calculated. 2. A method for classifying a woven fabric into a standard woven fabric based on standard data of the woven fabric, wherein the woven fabric to be classified and the standard woven fabric are represented by a confidence value for each characteristic element of the woven fabric. Data is calculated from the standard data, and the calculated value of the characteristic data of the classification target fabric and the standard fabric is arranged in an n-dimensional space, and the closest distance to the characteristic data value of the classification target fabric in the space. A method for classifying fabrics, characterized in that a standard fabric having a characteristic data value in 1 is selected as a similar fabric. 3. A device for allocating and classifying a woven fabric to a standard woven fabric based on the woven fabric standard data, a standard data storing means for storing the woven fabric to be classified and the standard data of the standard woven fabric, and each characteristic element of the woven fabric. Characteristic data represented by a confidence value regarding the standard data, characteristic data storage means for storing the characteristic data of the classification target fabric and the standard fabric calculated by the characteristic data calculation device, A fabric classifying apparatus comprising: a similar fabric searching unit that searches the standard fabric most similar to the classification target fabric from the feature data stored in the data storage unit. 4. The similar fabric searching means arranges the values of the feature data of the classification target fabric and the standard fabric in an n-dimensional space, and sets the feature data value of the classification target fabric in the space to the closest distance. The fabric classification device according to claim 3, wherein a standard fabric having a certain characteristic data value is selected as a similar fabric.