JP2003342869A - Method for controlling dyeing speak of synthetic fiber - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a method for controlling dyeing speck of a synthetic fiber to enable the use of information on a single fiber group constituting a yarn (monofilament, ultrafine fiber, modified cross-section fiber, hollow fiber, or the like) as the information for the investigation of the dyeing speck from the abnormality at the production step of the fiber and objectively and automatically evaluate and/or estimate the generation of dyeing speck without dyeing a knit or woven fabric and functionally evaluating the speck by the naked eyes. <P>SOLUTION: The cross section of a synthetic fiber is taken in as an image data, the data is processed and compared with a normal cross section, the discrepancy of the observed cross section from the normal cross section is analyzed, the generation of dyeing speck of the synthetic fiber is estimated from the analyzed discrepancy and the dyeing speck of the synthetic fiber is controlled by the estimation result. <P>COPYRIGHT: (C)2004,JPO

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for controlling uneven dyeing of synthetic fibers such as polyester, polyolefin and polyamide (in particular, ultrafine fibers, modified cross-section fibers or hollow fibers).

[0002]

2. Description of the Related Art In synthetic fibers such as polyester and polyamide, in recent years, modified cross-section fibers and hollow fibers having added value such as gloss, dry touch and light weight have been widely produced. Woven and knitted fabrics using these modified cross-section fibers and hollow fibers have distinctive luster, texture, and lightness compared to solid round cross-section fibers, and have become popular for clothing applications. .

A major problem with such modified cross-section fibers and hollow fibers is that spots are partially generated during dyeing, and dyeing varies depending on the spinning date and time. is there. These have a monofilament diameter of several tens of μ
It is considered that the shape of the discharge hole of the die has a complicated shape in order to form a modified cross section or a hollow at m, and the subtle changes in the discharge situation cause abnormalities in fineness and fiber shape. .

Generally, in order to detect abnormalities in irregularly shaped cross-section fibers and hollow fibers, an enlarged photograph of the fiber cross section of a yarn sampled from a wound yarn package is taken and the abnormality is detected from this cross-section photograph. The method of human monitoring is adopted. In addition, the method of estimating and evaluating the dyeing density of the fiber is to sample the fiber from the wound yarn package, create a knitted fabric, dye this, and classify the dyeing intensity by visual inspection. Is the actual situation. Therefore, there are many problems that the evaluation takes time, labor is required, and the visual judgment, that is, the sensory test is performed, which causes an erroneous judgment with an artificial judgment error.

Therefore, in the case of estimating and evaluating the dyeing density of a fiber, an example will be described. Instead of the visual inspection of the dyeing density, in which a knitted fabric is prepared and dyed as described above, for example, Japanese Patent Laid-Open No. 9-68466 or Japanese Patent Laid-Open No.
In Japanese Patent Application Laid-Open No. 0-253453, there is proposed a technique of irradiating a fiber before dyeing with near-infrared rays and obtaining a dyeing density from the absorbance or reflectance of the near-infrared rays.

In this prior art, first, the dyeing intensity of the fiber is not directly measured from the spectrum obtained by irradiating the near infrared ray, but the dyeing intensity after dyeing and the absorbance (or transmittance) of the near infrared ray are preliminarily dyed. ) Or a method of obtaining through a correlation regression equation with reflectance.

However, it is difficult for these prior arts to accurately measure on-line for a yarn running at a high speed, a yarn having a small fineness such as an ultrafine yarn, and a monofilament. Further, for a yarn having an irregular cross-sectional shape or a hollow shape, since the absorbance and reflectance of near infrared rays are irregularly reflected due to the peculiar shape, the disturbance factor is large, and the amount of reflected light is also Due to the small number, it is difficult to perform online measurement with sufficient accuracy. Therefore, it is actually very difficult to directly measure the dyeing density online from the running yarn itself.

Therefore, instead of measuring the dyeing depth on-line from the running yarn itself, near-infrared rays are applied to the yarn package in a wound state in which the yarns are accumulated, and its absorbance and reflectance are measured. Measuring is a realistic measuring method. Because these techniques require increasing the amount of reflected light to improve the accuracy of the measurement, they have no choice but to target a fiber assembly having a form such as a yarn package. It is limited to predicting the dyeing density using the information of the yarn assembly as a bulk.

Moreover, the above-mentioned prior art is specialized in the evaluation of dyeing density, and moreover, it is impossible to obtain information such as the shape change with time which the single fibers themselves constituting the yarn receive. Information that is impossible and has a change in the hollow ratio, a change in the degree of irregularity, etc. possessed by the individual single fibers constituting the yarn is not essentially utilized. For this reason, although it is possible to predict the dyeing density of the fiber, it is extremely difficult to determine the cause when an abnormality relating to the dyeing density occurs.

[0010]

SUMMARY OF THE INVENTION An object of the present invention is to solve the above-mentioned problems of the prior art. That is, the information (monofilament, ultrafineness, modified cross-section fiber, hollow fiber, etc.) from the single fiber group that constitutes the yarn can be used as information for tracing back to abnormalities in the fiber manufacturing process. Deterioration control of synthetic fibers that can objectively and automatically evaluate and / or predict the occurrence of stains without taking the method of actually dyeing the knitted fabric after the production of the knitting and sensually evaluating the abnormalities To provide a method.

[0011]

Means for Solving the Problems The present inventor has taken in image processing means image data obtained by photographing a cross section of a yarn sample taken from a wound yarn package by a known method. It is possible to evaluate and / or predict fiber stains by analyzing the captured image data and automatically analyzing it to analyze changes in the cross-sectional shape of the single fibers that make up the yarn that have changed over time. And found the present invention.

That is, according to the present invention, "a cross-sectional shape of a synthetic fiber is captured as image data, and the captured image data is image-processed and then compared with a normal cross-sectional shape,
Synthetic fiber stain management method for analyzing a change from a normal cross-sectional shape, predicting the occurrence of spots on the synthetic fiber from the analyzed changes, and managing the spots on the synthetic fiber by predicting the occurrence of spots Is provided.

At this time, when the synthetic fiber is a hollow fiber, the change from the normal cross-sectional shape is obtained by obtaining the area ratio between the hollow portion and the non-hollow portion of the hollow fiber, and the hollow portion is calculated by the area ratio. It is preferable to manage the dye spots on the fiber. At this time, the area value of the hollow portion and / or the non-hollow portion is calculated from the captured image data of the cross-sectional shape, not only for controlling spot stains but also for controlling the process and / or the product, and the area value is It is preferable to discriminate that the product is not in a normal value range and identify whether the product is abnormal or not to manage the product.

Further, when the synthetic fiber is a modified cross-section fiber, the degree of deformation is calculated from the cross-sectional shape of the modified cross-section fiber, and the change of the calculated variation from the normal degree of deformation is analyzed. It is preferable to manage the occurrence of stains on synthetic fibers by predicting the occurrence of stains.

At this time, the internal region of each single fiber is individually and separately recognized by the labeling process from the image data consisting of the cross-sectional shape of the single fiber group constituting the synthetic fiber to determine the contour of the internal region, and the separation and recognition are performed. The center of gravity of each monofilament is obtained from each of the inside regions, the distance from the obtained center of gravity to the contour is detected at each contour point existing on the contour, and the shape is deformed according to the ratio of the detected maximum distance and the minimum distance. It is preferable to evaluate the degree.

In particular, in the case where the modified cross-section fiber is a synthetic fiber having a tri-lobal type or triangular rice ball-shaped modified cross section, the three longest distances and three of the calculated distances from the center of gravity to the contour are used. A circumscribed circle and an inscribed circle respectively defined by the three vertices and the three bottom points on the contour line that form the shortest distance are calculated, and the shape is changed according to the ratio between the calculated diameter of the circumscribed circle and the diameter of the inscribed circle. It is preferable to evaluate the degree.

[0017]

BEST MODE FOR CARRYING OUT THE INVENTION According to the method of the present invention, "after the image data of the cross section of the yarn sample taken from the wound yarn package, which is taken by a known method, is taken into the image processing means, By automatically analyzing the captured image data and analyzing the changes in the cross-sectional shape of the single fibers that make up the yarn that have changed over time, it is possible to evaluate and / or predict fiber stains. " A major feature. Therefore, this point will first be described in detail with reference to the drawings.

FIG. 1 is a diagram showing the relationship between the hollow ratio of hollow fibers having a triangular cross section and the dyeing of the fibers. In this figure, the horizontal axis indicates the number of days that have elapsed since the spinneret was replaced and the spinneret was started up, and the spinneret surface was cleaned during startup.The vertical axis indicates the hollow section and non-hollow section from the cross-sectional shape of the hollow fiber. The graph of the hollow ratio in which the area ratio of the parts is displayed in percentage and the graph of the dyeing are overlapped and displayed.

As can be seen from this figure, as the hollow ratio changes as the number of spinning days elapses, at the same time, a difference appears in dyeing. Moreover, it can be seen that there is an extremely deep correlation between the decrease in hollow ratio and the deterioration in dyeing. Therefore, by monitoring the hollow ratio exemplified as one parameter expressing the change in the cross-sectional shape of the fiber, and detecting the change from its normal value (the hollow ratio of 9.4% in the figure is the normal value). It can be seen that stains can be evaluated and / or predicted, which makes it possible to control the dyeing of synthetic fibers. Although not described in detail, the same thing can be said by observing the change in the degree of irregularity in the irregular cross-section fiber.

Regarding "dyeing" shown on the vertical axis in FIG. 1, each of the obtained samples was knitted, and then, as a disperse dye, Terraseal Blue GF.
L: Dyeing condition using 2% owf 100 ℃ × 40 minutes
-Bath ratio of 1: 100 was used for dyeing, which was evaluated by a sensory test by comparing with a reference sample as an index of dyeing density.

The embodiments of the present invention will be described below in detail with reference to the drawings. FIG. 2 shows an example of an image showing a cross-sectional shape of a hollow fiber having a triangular cross section used for carrying out the present invention, and FIG. 3 shows a flow of image analysis processing used for carrying out the present invention. It is a flow.

FIG. 2 is an image obtained by bundling hollow monofilament fibers and cutting the fibers at the same time, and the cut surface (transverse cross section) thereof is photographed by an optical microscope. In this figure, the white portion indicated by reference numeral 2 is a non-hollow portion (fiber portion), and the black portion surrounded by the fiber portion 2 and indicated by reference numeral 3 is a hollow portion.

In the present invention, a commercially available personal computer and software are used to perform appropriate image processing on the captured image, analyze the cross-sectional shape of the synthetic fiber, and analyze the change from the normal cross-sectional shape. Analyzed. At this time, the central processing unit (CPU) used for image processing was 6
A personal computer with a processing capacity of 40 MHz and a memory capacity of 256 MBytes is used. The software is "Visual Basic Ver.6" from Microsoft and "LEADTOOLS Pro" from Bunka Orient, which is an image processing tool kit. The image processing software was used.

The process flow will be described in detail below with reference to FIGS. 2 and 3.

First, the image data having the cross-sectional shape of the fiber illustrated in FIG. 2 (a) is loaded into a personal computer for image processing by a known method. Regarding the acquisition of this image data, it is possible to use a charge-coupled device (CCD) camera to directly import it into the computer, or it is possible to read it once with a scanner from photographic printing paper taken with an ordinary optical camera and import it into the computer. Is.

Then, after the image data is thus taken into the personal computer, for example, a binarization process which is a known image processing method is performed, and as shown in the embodiment example of FIG. The part 2 and the background 1 and the hollow part 3 are clarified and separated, and by analyzing the information obtained in this way, a system for managing the unevenness of the synthetic fiber is constituted.

The system will be described below in accordance with the procedure shown in the flow chart of FIG.

First, a pre-processing 10 is performed in order to perform a series of image processing. The pre-processing 10 is binarized as long as it is not necessary to perform complicated processing because of its easy handling. Since it is preferable to use the processing as image processing, preprocessing for performing the binarization processing is usually used. As the preprocessing 10 for performing such binarization processing, processing such as clarification of an image to be processed, noise removal, or feature extraction as necessary is usually performed.

That is, in this example, the fiber portion 2 is "white" and the non-fiber portion such as the hollow portion 3 is "black", so that "gray" which is an intermediate color between "white" and "black" is not interposed. Binarization processing is performed. At that time, in order to remove noise or make the image clearer in performing the binarization processing, if necessary, for example, brightness, contrast, γ correction, sharpness, spatial filter, A process for obtaining an image that is easily binarized by using a known image correction method such as a negative or median filter is performed as the pre-processing 10.

In the image correction method, if the processing procedure is determined for the same sample group, the correction procedure will be the same. Therefore, if the above-described processing procedures are classified and programmed for each same sample and automatically processed by a personal computer according to the program, efficient processing can be performed when there are many measurement images. Needless to say.

Next, the binarization process is performed subsequent to the above-described pre-processing 10, and this binarization process is, as is known, 2
This is a process of setting the binarization level and determining whether to replace black or white at that level. Therefore, in order to accurately and surely binarize the binarized image, the obtained cross-section image (section image) is used to binarize Considering that the binarization levels naturally differ, it is necessary to set the binarization level value in advance by experiments.

In this way, the fiber portion 2 and the hollow portion 3 constituting the single fiber are clarified by the binarization treatment, and then the separation treatment 12 is performed. As the separation processing 12 as described above, a known labeling processing method which is a general method of image processing as shown in FIG. 4 is used in the present invention.

This labeling process is performed by binarizing a target region to be measured and then separately recognizing a target to be detected. In this case, a human being "has a number of objects of such a shape and of this size. Can be easily determined. But,
It is the result of growing the neural network in the growing period from infancy to the present. However, in the mechanical processing, the machine can only recognize that there are a plurality of points, and therefore it is necessary to recognize the above-mentioned flow path portion as a connected aggregate of connected points.

For this reason, first, the captured image data is coordinated in order to specify the position of each pixel forming the image data. Then, the whole area (all pixels) of the coordinated image is scanned to find a white pixel which is not yet labeled (not marked) as a flow path portion, and when such a white pixel is found, another white pixel is found. In this processing, a new label is added to distinguish it from the flow path portion, and white pixels are searched for in the adjacent pixels on the upper, lower, left, and right of the found white pixel to perform labeling.

The labeling process will be described more specifically. As shown in FIG. 4, the computer recognizes the internal region (fiber portion 2) constituting one single fiber as a connected region (connected aggregate). This is the process to be performed. That is,
In this example, the fiber portion 2 constituting each individual fiber is separated from the background 1 and the hollow portion 3 in its contour, and a process for recognizing the internal region (fiber portion 2) is performed. Is assigned to each. In the example of FIG. 4, the internal regions (connecting aggregates) of individual monofilaments are individually recognized by labeling with the numbers “1”, “2”, and “3”.

Next, the cross-sectional shape of the synthetic fiber is analyzed from the image separated into the fiber portion, the hollow portion, and the background portion by the separation processing 12, and for example, in the case of a hollow fiber, the hollow portion and the fiber portion are analyzed. The area is calculated, the hollow ratio is calculated to calculate the area ratio, and if the fiber has a modified cross-section, various parameters that characterize the cross-sectional shape such as the degree of modification 1 are calculated.
Do 3. Here, hollow ratio = (area of hollow portion 2)
/ (Area of the fiber portion 3), and the degree of irregularity will be described later.

Calculation processing 13 of such various parameters
In, for example, the area calculation is obtained from the number of pixels (the number of dots) included in the internal region of the fiber portion 2 or the hollow portion 3 assigned by the labeling described above. At this time, it is necessary to measure the number of dots of 1 mm in actual size in advance. Further, as a matter of course, the hollow ratio is performed for all the single yarns in the section image.

At this time, various statistical processing methods may be used in combination, such as obtaining an average value or a deviation from the data group obtained by the hollowness calculation. In addition, the area of the hollow portion 2, the area of the fiber portion 3, the hollowness, and the values obtained by statistical processing, etc., can be applied to the management of hollow fiber products, etc., if they are written into spreadsheet software or the like. It is preferable because it is possible.

By the way, when the single fiber group constituting the synthetic fiber is a modified cross-section fiber, it is preferable to perform the calculation process 13 of various parameters which characterize the cross-sectional shape as follows. Will be described in detail with reference to.

As described above, the inner region of each single fiber is individually separated and recognized by the labeling process from the image data having the cross-sectional shape, and the contour and inner region of the fiber portion of each single fiber are determined. At that time, the center of gravity of each single fiber is obtained from each separately recognized internal region by the following formula. However, the captured image data is coordinated, and the position (Xi coordinate, Yi coordinate) of each pixel forming the fiber portion 2 is (Xi, Yi) (i = 0.
~ N-1, where n is the number of pixels.

[0041]

[Equation 1]

Then, the distance from the center of gravity thus obtained to the contour is detected at each contour point existing on the contour, and the irregularity is evaluated by the ratio of the detected maximum distance and the minimum distance. is there.

By the way, when the synthetic fiber to be evaluated is a modified cross-section fiber, especially in the case of a synthetic fiber having a so-called triangular cross section such as a tri-lobal shape or a triangular rice ball shape, the method shown in FIG. , Of the distances from the center of gravity to the contour, the three vertices (three points farthest from the center of gravity) and the three bottom points (closest to the center of gravity) on the contour line that form the three longest distances and the three shortest distances. 3 points) and the circumscribed circle and the inscribed circle respectively defined by
The degree of irregularity can be evaluated by the calculated ratio of the diameter of the inscribed circle and the diameter of the circumscribed circle.

The above-described method of the present invention uses a circular hollow fiber having a triangular cross section as an example, but the present invention is not limited to such an example, and other fibers having different cross-sectional shapes and a large number of hollows are provided. It is clear from the gist that it can be widely applied to all hollow fibers such as the above-mentioned fibers.

[0045]

As described above, the present invention is a method for controlling stain spots by determining the ratio of the hollow part to the fiber part of the hollow fiber, the area of the hollow part or the area of the fiber part. is there.

This method clarified what should be controlled regarding the problem of dyeing occurring in the hollow fiber. Further, as the apparatus, image processing of the section image of the fiber is performed, so that a large amount of data can be easily collected. Therefore, defective products with abnormal dyeing can be prevented from entering the market, and the cause of the abnormalities can be identified quickly and easily, so cleaning of the spinneret surface and changing spinning conditions to normal conditions can be performed. It has the advantage that you can immediately take measures such as returning to.

[Brief description of drawings]

FIG. 1 is a view exemplifying a correlation between hollowness and dyeing, which is one of the major features in applying the method of the present invention.

FIG. 2 is an explanatory diagram illustrating (a) raw image data to be captured and (b) binarized image data used in the present invention.

FIG. 3 is a flowchart illustrating an outline of processing of the present invention.

FIG. 4 is an explanatory diagram for schematically explaining labeling processing used in the present invention.

FIG. 5 is an explanatory view schematically illustrating a method of obtaining a circumscribed circle and an inscribed circle for defining the degree of irregularity in a modified cross-section fiber.

[Explanation of symbols]

1 section image 2 fiber parts 3 Hollow part 10 Pretreatment 11 Binarization processing 12 Separation processing 13 Processing of various parameters

[Procedure amendment]

[Submission date] December 24, 2002 (2002.12.
24)

[Procedure Amendment 1]

[Document name to be amended] Statement

[Name of item to be corrected] 0032

[Correction method] Change

[Correction content]

In this way, the fiber portion 2 and the hollow portion 3 constituting the single fiber are clarified by the binarization treatment, and then the separation treatment 12 is performed. As such separation processing 12, a known labeling processing method which is a general method of image processing as shown in FIG. 4 is used in the present invention.

[Procedure Amendment 2]

[Document name to be amended] Statement

[Correction target item name] 0033

[Correction method] Change

[Correction content]

This labeling process is performed by binarizing a target region to be measured and then separately recognizing a target to be detected. In this case, a human being "has a number of objects of such a shape and of this size. Can be easily determined. But,
It is the result of growing the neural network in the growing period from infancy to the present. However, in machine processing, the machine can only recognize that there are a plurality of points, so it is necessary to recognize the above-mentioned fiber portion as a connected aggregate of connected points.

[Procedure 3]

[Document name to be amended] Statement

[Correction target item name] 0034

[Correction method] Change

[Correction content]

For this reason, first, the captured image data is coordinated in order to specify the position of each pixel forming the image data. Then, the entire area (all pixels) of the coordinated image is scanned to find a white pixel that is not yet labeled (not marked) as a fiber portion , and if such a white pixel is found, another fiber is detected. In this process, a new label is added to distinguish it from a copy, and white pixels are searched for in the adjacent pixels on the top, bottom, left, and right of the found white pixel, and labeling is performed.

[Procedure amendment 4]

[Document name to be amended] Statement

[Correction target item name] 0035

[Correction method] Change

[Correction content]

The labeling process will be described more specifically. As shown in FIG. 4, the computer recognizes the internal region (fiber portion 2) constituting one single fiber as a connected region (connected aggregate). This is the process to be performed. That is,
In this example, the fiber portion 2 constituting each individual filament is separated from the background 1 and the hollow portion 3 in its contour, and a process for recognizing the internal region (fiber portion 2) is performed to obtain the fiber of each individual filament.
A process of allocating each part 2 is performed. In the example of FIG. 4, the internal regions (connecting aggregates) of individual monofilaments are individually recognized by labeling with the numbers “1”, “2”, and “3”.

[Procedure Amendment 5]

[Document name to be amended] Statement

[Correction target item name] 0036

[Correction method] Change

[Correction content]

Next, the cross-sectional shape of the synthetic fiber is analyzed from the image separated into the fiber portion, the hollow portion, and the background portion by the separation processing 12, and for example, in the case of a hollow fiber, the hollow portion and the fiber portion are analyzed. The area is calculated, the hollow ratio is calculated to calculate the area ratio, and if the fiber has a modified cross-section, various parameters that characterize the cross-sectional shape such as the degree of modification 1 are calculated.
Do 3. Here, hollow ratio = (area of hollow portion 3 )
/ (Area of the fiber portion 2 ), and the degree of irregularity will be described later.

[Procedure correction 6]

[Document name to be amended] Statement

[Correction target item name] 0038

[Correction method] Change

[Correction content]

At this time, various statistical processing methods may be used in combination, such as obtaining an average value or a deviation from the data group obtained by the hollowness calculation. In addition, the area of the hollow portion 3, the area of the fiber portion 2 , the hollowness, and the values obtained by the statistical processing can be applied to the management of the hollow fiber products by performing a processing for writing them into spreadsheet software or the like. It is preferable because it is possible.

Continued front page    F term (reference) 2F065 AA17 AA52 AA58 CC00 FF04                       JJ03 JJ26 PP24 QQ04 QQ25                       QQ26 QQ31 RR08                 2G051 AA44 AB02 AB12                 3B154 AA06 AB01 AB03 AB04 AB05                       AB06 AB11 BA53 BB77 BC50                       CA02 CA23 CA27 DA13 DA30

Claims (6)

[Claims] 1. A cross-sectional shape of a synthetic fiber is captured as image data, the captured image data is subjected to image processing and then compared with a normal cross-sectional shape, and a change from the normal cross-sectional shape is analyzed and analyzed. A synthetic fiber stain control method for predicting the occurrence of stain spots on the synthetic fiber based on the change, and managing the stain spots on the synthetic fiber by predicting the occurrence of stain spots. 2. The synthetic fiber is a hollow fiber, the area ratio between the hollow portion and the non-hollow portion of the hollow fiber is calculated, and the unevenness of the hollow fiber is controlled by the area ratio. Management method of stains on synthetic fibers. 3. The abnormal product is managed by determining the area of the hollow portion and / or the non-hollow portion from the captured image data of the cross-sectional shape and determining that the value is not within a normal value range. 2. The synthetic fiber stain control method described in 2. 4. The synthetic fiber is a modified cross-section fiber,
The synthetic fiber stain control method according to claim 1, wherein the irregularity degree is calculated from the cross-sectional shape of the irregularly shaped cross-section fiber, and the change in the calculated irregularity degree from the normal irregularity degree is analyzed to predict the occurrence of stain spots. . 5. From the image data consisting of the cross-sectional shape of the single fiber group constituting the synthetic fiber, the inner region of each single fiber is individually and separately recognized by the labeling process to determine the contour of the inner region, and the separately recognized. The center of gravity of each monofilament is obtained from each internal region, the distance from the obtained center of gravity to the contour is detected at each contour point existing on the contour, and the degree of irregularity is determined by the ratio of the detected maximum distance and the minimum distance. The method for managing stains on hollow fibers according to claim 4, wherein 6. The modified cross-section fiber is a synthetic fiber having a tri-lobal or triangular diaper shaped modified cross-section, wherein the distance from the center of gravity to the contour detected by the method according to claim 5. Among them, the circumscribed circle and the inscribed circle respectively defined by the three vertices and the three bottom points on the contour line that constitute the three longest distances and the three shortest distances are calculated, and the calculated diameter of the circumscribed circle is calculated. The method for managing unevenness of hollow fiber according to claim 4, wherein the degree of irregularity is evaluated by the ratio with the diameter of the inscribed circle.