JPH11326238A - Apparatus for detecting defect of sheet and formation for sheet - Google Patents

Abstract

PROBLEM TO BE SOLVED: To highly accurately detect defects of even a sheet woven irregularly, by arranging sheet polarizers between the sheet to be measured and a light source, a photodetecting part with specific angles in direction of polarization and detecting bright, dark defects on the basis of binarized data of image signals. SOLUTION: This apparatus has a light source 1 and a photodetecting part 3 with a non-woven cloth 4 held therebetween. Sheet polarizers 2a, 2b are arranged between the non-woven cloth 4 and the light source 1 and between the non-woven cloth 4 and the photodetecting part 3 with approximately 70-110 deg. of directions of polarization. A light from the light source 1 passes the sheet polarizer 2a, non-woven cloth 4 and sheet polarizer 2b to be detected at the photodetecting part 3. The detected light is binarized by a binarization process means 6a to larger, smaller signal values (a), (b) (a>b) than a first threshold value, and larger, smaller signals than a second threshold value smaller than the first threshold value are binarized at a binarization process means 6b to values (b), (c) (b>c). The binarized signals are detected by a defect detection means 7 as bright defects if showing the value (a) to an input signal or as dark defects is showing the value (c). The detected defect is marked by 11 at an end part of the non-woven cloth 4.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an apparatus suitable for in-line detecting defects of a sheet such as a nonwoven fabric and a method for producing a sheet using the apparatus.

[0002]

2. Description of the Related Art When manufacturing a sheet having optical transparency and optical anisotropy, for example, a non-woven fabric, particularly in a process such as stretching or spreading, a pinhole defect or a local concentration of fibers occurs. As a result, an abnormal film thickness may occur, a drip defect may occur due to the incorporation of the molten fiber mass, and an undrawn yarn defect may occur in which undrawn fibers may enter. In order to prevent non-woven fabrics with such defects from being shipped as products, and to correct defects in the manufacturing process immediately when defects occur so as not to produce defective products, It is necessary to reliably detect inline. Further, in the nonwoven fabric, since the occurrence of unevenness of the fabric is inevitable even if it does not cause a problem as a product, it is necessary to detect a defect which is not affected by the unevenness of the fabric.

Conventionally, the detection of defects in nonwoven fabrics has relied on visual observation. However, it is extremely difficult to detect the defects only by instantaneously distinguishing the formation of the nonwoven fabric running at high speed from the defects. Since it is rate-limiting and it is difficult to improve productivity, automation using image processing technology has recently been considered. For example, Japanese Patent Application Laid-Open No. 6-18445
In the publication, a light source and a light receiving portion are arranged with a non-woven fabric interposed therebetween, and a polarizing plate is arranged between the non-woven fabric and the light source and between the non-woven fabric and the light receiving portion such that the polarization directions thereof are 20 degrees or less with each other. An apparatus for detecting a pinhole defect generated in a nonwoven fabric is described.

[0004] However, this conventional apparatus has a problem in that the pinhole defect, the abnormal film thickness defect, and the drip defect of the nonwoven fabric having a non-uniform weight are low in detection accuracy. In other words, regarding the pinhole defect, since the polarizing plates are arranged so that the polarization directions thereof are 20 degrees or less with each other, it is necessary to detect only the component in which the direction of the polarization plane has not changed during the transmission through the nonwoven fabric. become. Therefore,
The low-weight portion has very little light absorption and the rotation of the polarization plane of light is small, so that it is very bright, while the high-weight portion has very low brightness. Further, since the pinhole defect is slightly brighter than the low-weight portion, the obtained signal intensity differs depending on the brightness of the formation at the place where the pinhole defect occurs even with the same pinhole defect. In addition, the obtained signal intensity differs depending on the brightness of the formation at the place where the film thickness abnormality defect and the drip defect occur similarly to the pinhole defect,
Accurate detection has been difficult.

[0005] Further, regarding the undrawn yarn defect, the signal of the undrawn yarn defect portion is smaller than that of the formation portion because the conventional apparatus weakens the light intensity in the process of transmitting through the portion having optical anisotropy. It was a little dark and very difficult to detect.

[0006]

SUMMARY OF THE INVENTION An object of the present invention is to solve the above-mentioned problems of the prior art, and to provide a pinhole defect, an abnormal film thickness defect, a drip defect, even for a sheet having a non-uniform weight. It is an object of the present invention to provide an apparatus capable of accurately detecting an undrawn yarn defect and a sheet manufacturing method using such an apparatus.

[0007]

According to the present invention, in order to achieve the above object, a light source and a light receiving section are arranged with a sheet to be measured interposed therebetween, and a video signal obtained by the light receiving section is processed by an image processing means. This is an apparatus for detecting a defect of a sheet by arranging a polarizing plate between the sheet to be measured and the light source and between the sheet to be measured and the light receiving section such that the polarization directions form an angle of 70 to 110 degrees with each other. And binarizes the video signal with the first threshold value in the image processing apparatus.
Binarizing means, second binarizing means for binarizing the video signal with a second threshold lower than the first threshold, first and second binarizing means,
A defect detecting means for detecting each of the light and dark defects based on the binarized data obtained by the value converting means is provided.

The sheet to be measured is optically transparent and optical
Any sheet that has anisotropy may be used.
Something like synthetic resin film, paper, glass fiber sheet
It is. Above all, span where undrawn yarn defects are likely to occur
Particularly preferred is application to nonwoven fabrics produced by the bond method.
No. Among non-woven fabrics, white and the basis weight is 100 g / m ^Two
The following are particularly preferred in terms of the amount of transmitted light. Light color
Light transmittance changes due to thickness fluctuations and polarization
It is easy to balance changes in the detected light amount due to
The smaller the amount, the more likely the unevenness of the weight is to be generated.
The effect is easy to appear. In addition, as a non-woven fabric, the coating of electric wires
It is particularly preferred to apply it to the material used. Electrical wire
When used as a covering material, the nonwoven fabric should be
Lit and wrap it around the wire with strong tension. This
Undrawn yarn and drip defects in the non-woven fabric
When it is growing, cracks occur there, and due to tension
The nonwoven is cut. Then cut the wire and rewind
The problem arises that it must be done. In the present invention,
Precisely remove undrawn yarn and drip defects that cause
It can detect and detect non-woven fabric in the wire manufacturing process.
Cutting can be prevented beforehand.

The difference between the polarization directions of the two polarizing plates is most preferably 90 degrees in order to remove the unevenness in the weight. As a light source, a fluorescent lamp that is turned on at a high frequency is preferable because it can irradiate light more uniformly in the width direction of the sheet and does not have temporal noise in a video signal. Although a CCD camera can be used as the light receiving unit, a line sensor CCD camera in which pixels are arranged in a line is preferable because the inspection can be performed without interruption in the sheet traveling direction. More preferably, a plurality of solid-state imaging devices in which pixels are linearly arranged are arranged in the running direction of the nonwoven fabric, and time integration for accumulating and moving charges one after another to adjacent solid-state imaging devices in synchronization with the running speed of the nonwoven fabric. The line sensor camera of the type can receive transmitted light reduced by the polarizing plate with high sensitivity.

[0010] In the present invention, it is preferable that the image processing apparatus includes averaging processing means for averaging video signals in the traveling direction of the seat.

It is also preferable that the image processing apparatus includes a threshold value calculating means for calculating a first threshold value and a second threshold value from a video signal obtained from at least one light reception of the light receiving section.

Further, it is preferable that the sheet to be measured is provided with marking means for giving a mark indicating the presence of a defect. The marking means is preferably a means for marking the end of the sheet.

Further, it is preferable that a defect history management means is provided.

Further, the sheet manufacturing method of the present invention is characterized in that the nonwoven fabric manufacturing process is managed by using the above-mentioned defect detecting device.

[0015]

FIG. 1 shows an apparatus in which a light source 1 and a light receiving section 3 are arranged with a non-woven fabric 4 as a sheet to be measured interposed therebetween, and a polarizing plate 2a is interposed between the non-woven fabric 4 and the light source 1.
The polarizing plate 2b is placed between the light receiving section 3 and the
They are arranged so as to form an angle in the range of up to 110 degrees. The light receiving section 3 is connected to two binarization processing means. First
A first binarization processing unit 6a for binarizing a signal larger than the threshold value, and a second binarization unit for binarizing a signal smaller than the second threshold value
The binarization processing means 6b is connected to the defect detection means 7 via the data bus 8. The first binarization processing unit 6a, the second binarization processing unit 6b, the defect detection unit 7, and the data bus 8 constitute the image processing device 5. Further, the defect detecting means 7 is connected to the marking means 11 and the defect history managing means 12.

The light emitted from the light source 1 passes through the first polarizing plate 2a, the nonwoven fabric 4, and the second polarizing plate 2b, and is received by the light receiving section 3. The received light is taken into the image processing device 5 as a one-dimensional video signal, and a signal larger than the first threshold is set to a by the first binarization processing unit 6a.
In addition, a signal smaller than the first threshold is binarized to b (a> b). Further, the one-dimensional video signal is processed by the second binarization processing means 6b so that the one-dimensional video signal has a value smaller than the first threshold value.
A signal smaller than the threshold is binarized to c, and a signal larger than the second threshold is binarized to b (b> c). The signals binarized by the first and second binarization means are input to the defect detection means 7,
The defect detecting means 7 detects a defect as a bright defect if the value is a with respect to the input signal, and detects a defect as a dark defect if the value is c with respect to the input signal. For the detected defect, the marking means 11 makes a mark on the end of the nonwoven fabric where the defect has occurred, and the defect history management means 12 manages the time, place, defect type, area and the like when the defect occurred.

Here, the first polarizing plate 2a, the nonwoven fabric 4, the second
The polarization direction of the light transmitted through the polarizing plate 2b will be described. Light transmitted through the polarizing plate 2a becomes linearly polarized light that vibrates only in a certain direction. When this linearly polarized light passes through the normal yarn in the nonwoven fabric 4 to be inspected, it becomes elliptically polarized light due to the birefringence of the normal yarn. Furthermore, this elliptically polarized light is
The light is irradiated onto a polarizing plate 2b arranged between the light receiving unit 3 and the polarizing direction so that the polarizing direction is orthogonal to the polarizing direction of the polarizing plate 2a.
Of the light applied to the polarizing plate 2b, only a component having the same direction as the polarization direction of the polarizing plate is transmitted and received by the light receiving unit 3.

On the other hand, when the nonwoven fabric 4 does not exist, the light transmitted through the first polarizing plate 2a is applied to the second polarizing plate 2b without changing the polarization direction. The light transmitted through the first polarizing plate 2a and the polarization direction of the second polarizing plate 2b are orthogonal to each other, so that the light transmitted through the first polarizing plate 2a can be transmitted through the second polarizing plate 2b. Can not. Therefore, when the nonwoven fabric 4 does not exist (for example, when there is a pinhole), a dark signal is obtained. Therefore, when the non-woven fabric is present in the light receiving unit 3, a brighter signal is obtained as compared with the case where no non-woven fabric is present.

Here, considering the high-weight portion and the low-weight portion of the nonwoven fabric 4, the high-weight portion has a larger thickness than the low-weight portion, and thus has a low transmittance. on the other hand,
The light transmitted through the high-weight portion passes through the thread more than the light transmitted through the low-weight portion, and is thus largely polarized, and contains many components the same as the polarization direction of the second polarizing plate 2b.

As described above, the high-weight portion has low light transmittance, but has many components in the same direction as the polarization direction of the second polarizing plate 2b. Conversely, the low-weight portion has high light transmittance. The component that is the same as the polarization direction of the second polarizing plate 2b is small. Therefore,
Since the light transmittance and the light passing through the second polarizing plate 2b can be balanced between the high and low weight portions,
The difference in the amount of transmitted light becomes smaller. As a result, if the weight is within the normal range, the difference in the amount of transmitted light due to the weight can be removed to a level that does not cause a problem, and a video signal with a uniform formation can be obtained. Further, the brightness of the formation of the video signal has a correlation with the basis weight set for each product.

The signal detected for each of the drawbacks will be described below.

As for the pinhole defect, the light transmitted through the polarizing plate 2a passes through the pinhole defect without changing the polarization direction and is irradiated on the polarizing plate 2b. Since the light transmitted through the polarizing plate 2a and the polarization direction of the polarizing plate 2b are orthogonal to each other, the light transmitted through the polarizing plate 2a and passing through the pinhole defect cannot be transmitted through the polarizing plate 2b. For this reason, with respect to the pinhole defect, a signal which is sufficiently darker than the video signal of the formation is obtained, and can be detected as a dark defect.

The abnormal film thickness defect and the drip defect are much higher in weight and thicker than those in the high-weight portion. Even if the light transmitted through this portion is polarized in a large amount, the transmitted light amount itself is extremely small, so that a signal sufficiently darker than the video signal of the formation is obtained and can be detected as a dark defect. However, the drip defect having a small film thickness can be detected as a bright defect, as in the case of the undrawn yarn described below, a signal sufficiently brighter than the video signal of the formation is obtained.

The birefringence of the undrawn yarn is larger than the birefringence of the normal yarn, and the light transmitted through the undrawn yarn becomes elliptically polarized light closer to circularly polarized light than the light transmitted through the normal yarn. Therefore, for the light transmitted through the undrawn yarn, the same component as the polarization direction of the polarizing plate 2b is larger than that of the normal yarn, so that a sufficiently bright signal can be obtained as compared with the video signal of formation, and Can be detected as

As described above, according to the defect detecting apparatus of the present invention, even in a nonwoven fabric having a non-uniform weight, a video signal of a uniform formation can be obtained, and various defects occurring in the nonwoven fabric can be sufficiently detected. Since a signal can be obtained, detection can be performed with high accuracy.

FIG. 2 shows an apparatus according to another embodiment of the present invention in which averaging means 9 is added to the image processing apparatus 5. When the nonwoven fabric 4 is embossed, light and dark stripes are included in the video signal obtained at the light receiving portion. However, even with such a nonwoven fabric, the stripes can be removed by averaging in the running direction. Defects can be detected accurately. In FIG. 2, the averaging processing means 9 is connected to the light receiving section 3 and two binarization processing means. Here, the averaging means 9 stores the inputted one-dimensional video signals in the buffer by the number corresponding to the length to be averaged, and averages the plurality of one-dimensional video signals in the running direction of the nonwoven fabric. Perform a conversion process. When a new one-dimensional video signal is input to the buffer, the averaged 1
A two-dimensional video signal is output and input to the first and second binarizing means.

FIG. 3 shows that the image processing apparatus 5 has a threshold calculating means 1.
9 shows an apparatus according to yet another embodiment of the present invention, to which 0 is added. For nonwoven fabrics that differ only in basis weight,
The basis weight has a correlation with the brightness of the formation of the video signal. For this reason, by setting the threshold value from the video signal, the trouble of setting the threshold value can be saved. In FIG. 3, the threshold value calculating means 10
Are connected to the averaging means 9 and the data bus 8. Here, the threshold value calculating means 10 calculates an average value of at least one one-dimensional video signal output from the averaging processing means 9, and a predetermined value for each type having a different basis weight from the average value, or A threshold value of the first and second binarizing means is calculated by adding, subtracting, multiplying, and dividing a value calculated from a variance value of the one-dimensional video signal. The calculated threshold is sent to the first and second binarizing means via the data bus 8. The calculation of the threshold value may be performed at regular intervals. Also, it is effective to perform this when replacing the lighting or when changing the product type. The averaging means 9 is not always necessary, but is preferably used because the defect can be detected more accurately.

[0028]

Example 1 A defect of a nonwoven fabric was detected using the apparatus shown in FIG. The light source 1 has an effective emission length of 550 mm and a frequency of 30 KH
A high frequency fluorescent lamp of z was used. As the light receiving unit 3, a time-integrating line sensor camera in which 96 solid-state imaging devices in which 2,048 pixels are arranged in the width direction of the nonwoven fabric 4 are arranged in the running direction of the nonwoven fabric 4 was used. The polarizing plate 2a and the polarizing plate 2b were arranged so that the polarization directions were orthogonal to each other. The nonwoven fabric 4 has a basis weight including a pinhole defect, an abnormal film thickness defect, a drip defect, and an undrawn yarn defect produced and spunbonded by a spun bond method and having a basis weight of 55 g /.
using the m ² of non-woven fabric. The running speed of the nonwoven fabric was 60 m / min.

As a result, it was possible to obtain a video signal having a uniform formation even for a nonwoven fabric having a non-uniform basis weight.

Regarding the pinhole defect, the abnormal film thickness defect, and the drip defect having a large film thickness, a sufficiently dark signal was obtained as compared with the formation signal, so that the signal was binarized by the second binarization means. As a dark defect, the defect detecting means could accurately detect the dark defect without being affected by the unevenness of the eyes.

Regarding the undrawn yarn defect and the drip defect having a small film thickness, a sufficiently bright signal can be obtained as compared with the formation signal, so that the first binarization means binarizes the signal.
As a bright defect, the defect detecting means was able to accurately detect a bright defect without being affected by unevenness in the weight. Example 2 A defect of the nonwoven fabric was detected using the apparatus shown in FIG.

The same light source, light receiving part, polarizing plate and non-woven fabric as the object to be inspected were the same as those used in Example 1.

As a result, the light and dark stripes due to embossing could be removed by the averaging means, so that pinhole defects, abnormal film thickness defects, drip defects and undrawn yarn defects can be detected more accurately. Was completed. Example 3 The defect of the nonwoven fabric was detected using the apparatus shown in FIG.

The same light source, light receiving section and polarizing plate as those used in Example 1 were used. As the nonwoven fabric, two types having a basis weight of 55 g / m ^{2 and} a fabric having a basis weight of 80 g / m ² were used. The calculation of the threshold value by the threshold value calculation means was performed from one video signal immediately after the start of the inspection.

As a result, the threshold value is calculated by adding, subtracting, multiplying, and dividing the average value of the one-dimensional video signal of each nonwoven fabric by the threshold value calculation means. Defects, abnormal film thickness defects,
All drip defects and undrawn yarn defects could be detected. Example 4 A defect of the nonwoven fabric was detected in-line by the apparatus shown in FIG.

The same light source, light receiving portion and polarizing plate as those used in Example 1 were used. As the non-woven fabric 4, an embossed fabric having a basis weight of 55 g / m ² by a spun bond method was used. The running speed of the nonwoven fabric is 60
m / min.

As a result, the generated defect could be detected with high accuracy, and when a defect occurred, a mark could be attached to the end of the nonwoven fabric where the defect occurred by the marking means. Also, even when defects occur frequently, the type, area,
Immediately identifying and correcting the defective part in the manufacturing process where the defect occurred from the place and time, it was possible to minimize the manufacture of defective products. Therefore, the manufacturing cost of the nonwoven fabric could be reduced.

When the non-woven fabric having no defect as a result of the inspection in the above example was slit into a 30 mm wide band and wound around the electric wire to produce an electric wire, the non-woven fabric could be cut during the winding. It was possible to increase the production efficiency of electric wires and reduce costs.

[0039]

According to the present invention, the polarizing plates are arranged between the sheet to be measured and the light source and between the sheet to be measured and the light receiving section so that the polarization directions are at an angle of 70 to 110 degrees with each other. Even with a sheet having uneven weight, unevenness of the formation of a video signal due to uneven weight can be reduced to a level that does not cause a problem, and a portion having a high birefringence is brighter than the formation, and the pin pole defect and It is possible to obtain a signal in which the film thickness defect is sufficiently darker than the formation.
Further, the image signal is transmitted to the image processing device by the first threshold value by 2.
A first binarizing unit for binarizing the video signal, a second binarizing unit for binarizing the video signal with a second threshold lower than the first threshold, and a first and a second binarizing unit. Since there is provided a defect detecting means for detecting each of the light and dark defects based on the obtained binarized data, the light and dark defects generated on the sheet can be detected. As a result, pinhole defects, abnormal film thickness defects, drip defects, and undrawn yarn defects, which occur in a sheet having uneven weight, can be accurately detected.

When the image processing apparatus has an averaging processing means for averaging the video signal in the running direction of the sheet to be measured, the image signal is periodically generated in a direction oblique to the running direction of the sheet. For example, light and dark stripes due to embossing or the like can be removed, and defects can be detected with higher accuracy.

In the case where the image processing apparatus is provided with threshold calculating means for calculating the first and second thresholds from video signals obtained from at least one light reception of the light receiving section, a plurality of products having different basis weights are provided. For this sheet, the trouble of setting the threshold value can be saved, and erroneous detection due to the setting error of the threshold value can be prevented.

In the case where the sheet to be measured is provided with marking means for giving a mark indicating the presence of a defect,
Even if the nonwoven fabric has a defect, the defect can be removed before shipping, and the yield can be improved.

Further, when the defect history management means is provided, the manufacturing process causing the defect can be immediately identified and treated, so that the production of defective products can be minimized and the production cost can be reduced. Will be able to

When the defect detecting device of the present invention is used to detect a defect of a non-woven fabric used as a covering material of an electric wire, an undrawn yarn or a drip defect that causes cutting of the non-woven fabric in the electric wire manufacturing process can be accurately detected. Can be detected.

Further, when the defect detecting device of the present invention is applied to a nonwoven fabric manufacturing process, a defective portion can be immediately corrected, and the manufacturing cost of the nonwoven fabric can be reduced.

[Brief description of the drawings]

FIG. 1 is a block diagram of a defect detection device according to an embodiment of the present invention.

FIG. 2 is a block diagram of a defect detection device according to another embodiment of the present invention.

FIG. 3 is a block diagram of a defect detection device according to still another embodiment of the present invention.

[Explanation of symbols]

 1: light source 2a: polarizing plate 2b: polarizing plate 3: light receiving portion 4: non-woven fabric 5: image processing device 6a: first binarizing means 6b: second binarizing means 7: defect detecting means 8: data bus 9: average processing means 10: threshold value calculation means 11: marking means 12: defect history management means

 ──────────────────────────────────────────────────の Continuing on the front page (72) Inventor Tetsuya Ito 1-1-1 Sonoyama, Otsu-shi, Shiga Pref.

Claims (8)

[Claims] 1. A light source and a light receiving section are disposed with a light transmitting and optically anisotropic sheet to be measured interposed therebetween, and a video signal obtained by the light receiving section is processed by image processing means. An apparatus for detecting a defect of a sheet to be measured, wherein a polarizing plate is disposed between the sheet to be measured and the light source and between the sheet to be measured and the light receiving unit so that the polarizing directions are 70 to 11 with respect to each other.
First binarizing means for binarizing the video signal with a first threshold value, the video signal being lower than the first threshold value; A second binarizing unit for binarizing with a second threshold value, and a defect detecting unit for detecting each defect of light and dark based on the binarized data obtained by the first and second binarizing units. A sheet defect detecting device, comprising: 2. The sheet defect detecting apparatus according to claim 1, wherein the image processing apparatus includes averaging processing means for averaging video signals in a running direction of the sheet to be measured. 3. The image processing apparatus according to claim 1, wherein the image processing apparatus includes a threshold value calculating unit that calculates first and second threshold values from a video signal obtained from at least one light reception of the light receiving unit. Sheet defect detection device. 4. The sheet defect detecting apparatus according to claim 1, further comprising a marking means for providing a mark indicating the presence of a defect on the sheet to be measured. 5. The apparatus according to claim 1, further comprising a defect history management unit.
5. The apparatus for detecting a defect of a sheet according to any one of claims 1 to 4. 6. The sheet to be measured is a non-woven fabric.
A sheet defect detecting apparatus according to any one of claims 5 to 10. 7. A sheet manufacturing method, wherein a sheet manufacturing process is managed by using the defect detecting device according to claim 1. 8. The nonwoven fabric is a covering material for an electric wire.
3. The method for producing a nonwoven fabric according to item 1.