JP2011058871A - Abnormality inspection device for spinneret and abnormality inspection method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an abnormality inspection device and an abnormality inspection method for spinnerets automatically detecting at least generation of bending of a pipe in tens of thousands of pipes attached in the inside of a spinneret for melt-spinning sea-island type conjugated fibers. <P>SOLUTION: The abnormality inspection device and the abnormality inspection for spinnerets detect the presence or absence of bending of a pipe in a spinneret to which a spinneret plate having a large number of straight-tube type pipes erected perpendicular to the spinneret plate is attached, by photographing inspection light having passed through through-holes of each pipe, which are polymer channels, by a camera; calculating the center coordinate of a polymer channel at the tip of each pipe by image processing such as binarization processing on the photographed image data; and determining whether the calculated center coordinate is deviated from a designed value. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a spinneret that melt-spins sea-island type composite fibers in which a number of islands made of island component polymers are formed in a sea component polymer along the fiber axis direction, and a number of trickled islands are formed in the spinneret. Abnormalities in the spinneret for automatically inspecting the foreign matter remaining in the pipes and other parts that are provided to discharge the component polymer into the sea component polymer, or abnormalities occurring in the spinneret such as deformation of the pipe itself The present invention relates to an inspection apparatus and an abnormality inspection method.

  A spinneret for producing a so-called “sea-island type composite fiber” in which a large number of island-component polymers are arranged in an island shape independently in the fiber axis direction in a sea-component polymer is known. For example, in Patent Documents 1 to 3 and the like, a large number of “pipes extending in a straight line (hereinafter also simply referred to as“ straight pipes ”)” that serve as the flow paths of the island component polymers are provided vertically, and There is disclosed a spinneret for sea-island type composite fibers having a configuration in which a sea component polymer is supplied so as to wrap around the outer peripheral portion of the base.

  A spinneret equipped with such a large number of straight pipes needs to be washed to remove the polymer remaining in the spinneret for reuse after being used for a certain period of time. However, as the island component polymer remaining in the straight pipe after use, for example, when a heat-resistant polymer that is difficult to thermally decompose or a polymer that has chemical resistance that is difficult to dissolve in a solvent is used, these polymers are completely removed. It may be difficult to do this. In such a case, it is difficult to produce sea-island type composite fibers having good quality and performance.

  Therefore, a spinneret that has been used once is made into a die that can be easily cleaned as a die structure that can be easily disassembled and assembled. However, in such a spinneret disassembly operation and assembly operation after cleaning, there is a problem that abnormalities such as pipe bending and damage are likely to occur in the straight pipe.

  Moreover, in recent years, in order to produce ultra-fine fibers, it has become necessary to increase the number of straight pipes as much as possible in order to increase the number of islands formed in one sea-island filament. As a result, the number of straight pipes provided in one sea-island type spinneret reaches tens of thousands (for example, Patent Document 3 describes an example in which 19,008 straight pipes are provided. ).

  However, if the polymer flow path of the straight pipe is dirty or foreign matter adheres in the polymer flow path, it may cause troubles that cause yarn breakage during spinning, making spinning impossible, or uniting of single yarns Or abnormal quality occurs. Of course, this situation also occurs when the straight pipe is damaged or bent.

  For the above-mentioned reasons, inspect the residual foreign matter in the straight tubular polymer flow path to confirm that there is no foreign matter, and also to confirm that there is no damage or bending of the straight pipe. Maintaining normality is the lifeline of spinning technology. For this purpose, it is generally carried out by carefully examining whether or not the straight polymer flow path is normally maintained by the inspector's naked eye using a microscope.

  Needless to say, such an inspection requires a lot of man-hours and time. Moreover, since it is a visual inspection, the work of inspecting for a long time requires the perseverance of the inspector, and the work is tired of eyes. Therefore, human errors sometimes tend to overlook the presence of anomalies. In addition, the labor cost is increased due to the need for an inspector, and furthermore, there is a problem that individual differences occur in the inspection.

  In addition, as an attempt to perform an automatic inspection without relying on a manpower for a cap inspection, there is one proposed in Patent Document 4, for example. However, this conventional technique inspects the presence or absence of foreign matter remaining in the discharge hole group for discharging the polymer from the spinneret, and inspects the bending of a large number of straight pipes provided in the spinneret. The necessity of such a function is not recognized at all.

  Needless to say, in the spinneret of the sea-island type composite fiber, the number of filaments to be spun is about 50 filaments (the aforementioned Patent Document 3 describes an example in which 48 filament groups are spun from one spinneret. Therefore, the number of discharge holes for spinning these filament groups is about 50 at most. For this reason, the burden for inspection of foreign matter remaining in the straight pipe and inspection of bending of the straight pipe and the like is far greater than expected than the inspection of foreign matter remaining in the discharge hole.

Japanese Patent Publication No. 44-18369 JP 2001-192924 A JP 2005-320640 A JP 2006-267000 A

  The object of the present invention is to detect foreign matters remaining in the polymer flow path in the tens of thousands of straight pipes installed in a die such as a sea-island type spinneret in view of the above-described problems of the prior art. An object of the present invention is to provide an abnormality inspection device and an abnormality inspection method for a spinneret that can automatically detect pipe deformation.

Here, as an invention for solving the above-mentioned problem, “a pipe produced in a pipe in a spinneret in which a large number of pipes having a circular channel cross-sectional shape extending in a straight line are vertically arranged with respect to a base plate” An abnormality inspection apparatus for a spinneret having the following requirements (1) to (3) for detecting at least a bend.
(1) Illumination for irradiating inspection light that passes through a through-hole that is a polymer flow path of the pipe vertically installed;
(2) A magnifying lens is provided to face the illumination with the base plate in between, and to magnify the inspection light irradiated from the illumination and passed through the group of through holes to a predetermined magnification. Camera
(3) The presence region of the polymer flow path at the pipe tip is specified from the image data obtained by photographing the inspection light that has passed through the through hole with the camera, and the center of the polymer flow path at the tip of the pipe from the specified existence region Calculate the coordinate value, and the calculated center coordinate of the polymer flow path at the tip of each pipe is more than the allowable value than the center coordinate value of the polymer flow path at the tip of each pipe determined at the time of designing the spinneret. An image processing apparatus having an identification function for identifying that pipe bending has occurred when it occurs is provided.

  At this time, in the spinneret abnormality inspection apparatus according to the present invention, the pipe is a pipe for discharging the island component polymer into the sea component polymer when melt spinning the sea island type composite fiber. It is preferable because it is possible to automatically inspect the bending of a pipe as long as possible.

Further, as an invention for solving the above-mentioned problem, “a pipe bend generated in the pipe in a spinneret in which a plurality of pipes having a circular flow path cross-sectional shape extending in a straight line are vertically arranged with respect to the base plate” A method for inspecting an abnormality of a spinneret having the following requirements (1) to (5) for detecting at least:
(1) Irradiate inspection light that passes through a through-hole that is a polymer flow path of the pipe installed vertically,
(2) The inspection light that has passed through the through hole is photographed as image data by enlarging the inspection light to a predetermined magnification at the pipe tip on the side from which the polymer is discharged,
(3) The image data is binarized with a gradation of light and dark with a predetermined threshold as a reference, and the pixel in which the inspection light passing through the through hole is detected is “bright”, and the inspection light is detected. The pixels that were not done are classified as “dark”
(4) Identify the existing region of the aggregate of pixels separated from "bright", calculate the center coordinate value of the polymer flow path at the tip of each pipe from the specified existing region,
(5) The calculated center coordinates of the polymer flow path at the tip of each pipe are uniquely determined at the time of designing the spinneret. If a pipe is bent, it is identified. Is provided.

  In this case, according to the present invention, it is possible to determine a threshold value for allocating the pixels constituting the image data to “bright” or “dark” by the P tile method in the binarization process, such as deterioration of a light source serving as illumination or a camera This is preferable because it is not affected by fluctuations in light and dark gradations of image data due to fluctuations in imaging conditions.

  According to the present invention described above, it is possible to inspect the abnormality of a large number of straight pipes provided in the spinneret, such as the spinneret of sea-island type fibers. In particular, these straight pipes can be easily and automatically inspected for the presence of pipe bending. In addition, as a matter of course, in many cases, it is possible to automatically perform an abnormality inspection of a straight pipe provided to reach tens of thousands in one spinneret without depending on the naked eye of the inspector.

It is front sectional drawing which illustrated typically the embodiment which described a mode that many straight pipes were provided in the spinneret which melt-spins a sea-island type composite fiber. It is a principal part enlarged view of the part shown with the dotted line in FIG. It is the perspective view which showed the example of embodiment of the inspection method of the pipe bending which can be used for this invention.

  In the spinneret abnormality inspection apparatus and abnormality inspection method according to the present invention, a large number of straight pipes (straight pipes) having a circular flow passage cross-sectional shape are vertically installed inside the spinneret. The main feature of the present invention is that it detects an abnormality in a spinneret such as a bent pipe that has occurred in at least one straight pipe.

  Hereinafter, embodiments of an abnormality inspection apparatus and an abnormality inspection method for a spinneret of the present invention will be described in detail with reference to the drawings. Before that, the abnormality inspection apparatus and the abnormality inspection method of the present invention are to be inspected. An embodiment of “a large number of straight pipes provided in the spinneret” will be specifically described with reference to FIGS. 1 and 2.

  FIG. 1 is a front sectional view exemplifying an embodiment schematically showing a state in which a large number of straight pipes are provided inside a spinneret, such as a spinneret of a sea-island type composite fiber. However, the left half of FIG. 1 has a cross section so that the flow path through which the island component polymer passes can be easily understood, and the right half of FIG. 1 has a cross section so that the flow path through which the sea component polymer passes can be easily understood. Each is shown.

FIG. 2 is an enlarged view of a main part of the portion indicated by a dotted line in FIG. 1, and illustrates “one constituent unit for spinning one sea-island type composite filament”. Normally, the structural unit for spinning such a single sea-island type composite filament is, for example, a spinneret for spinning 48 sea-island type multi-filament yarns as described in Patent Document 3 above. 48 structural units, for example, 8 and 16, respectively, on multiple concentric circles having the center of the base shown in FIG. 1 as a common center and three pitch circle diameters D ₁ , D ₂ , D ₃ , 24 (total 48) are provided in a uniform manner.

  In the “spinneret abnormality inspection apparatus and abnormality inspection method” of the present invention, when determining the presence or absence of abnormality of the spinneret, the above-mentioned “one constituent unit for spinning one sea-island type composite filament” It is preferable to carry out every time. This is because the straight pipes are grouped and densely arranged for each “structural unit” and are not arranged at other places. Therefore, it is because it is much more efficient to perform image processing for each such “structural unit” and the inspection accuracy is further improved.

  Here, the elements indicated by the respective symbols in FIGS. 1 and 2 will be briefly described. 1 is a spinneret, 11a to 11f are first to sixth base plates, 12 is a straight pipe, and 13 is an island component polymer introduced. The hole 14 is a sea component polymer introduction hole, 15 is an island component polymer reservoir, 16 is a sea component polymer reservoir, 17 is a polymer merge hole, 18 is a funnel-shaped channel, and 19 is a discharge hole.

  In the spinneret configured as described above, as shown in FIG. 1, the island component polymer is introduced into the island component polymer introduction hole 13 and then shunted to form the island component polymer reservoirs 15 constituting the respective structural units. Reach each. The other sea component polymer is introduced into the sea component polymer introduction hole 14 and then divided to reach the sea component polymer reservoir 16 constituting a part of each of the structural units.

  Next, the island component polymer that has reached each constituent unit in this way is diverted to the respective straight pipes 12 and discharged from the respective straight pipes 12 to the sea component polymer reservoir 16, whereby the island component that becomes the core is formed. A large number of core-sheath streams are formed around the polymer stream, and the sea component polymer that forms the sheath is formed. And this many core-sheath flow flows down the polymer confluence | merging hole as it is, mutually merges above the funnel-shaped flow path 18, and unites, and forms a sea-island type composite flow. Thereafter, the sea-island type composite flow is gradually narrowed as it flows down the funnel-shaped flow path 18, reaches the discharge holes 19, and is spun out as sea-island type composite fibers from the discharge holes 19.

  In the “abnormality inspection apparatus and abnormality inspection method” of the present invention relating to the spinneret described above, the presence or absence of foreign matter remaining in the discharge hole 19 can be inspected. The main feature is the detection of the presence or absence of residual foreign matter in the tube, which will be described in detail below with reference to FIG.

  In the “spinning cap abnormality inspection apparatus” of the present invention illustrated in FIG. 3, the straight tube 12 mounted on the fourth cap plate 11 d of the spinning cap 1 is irradiated with illumination 23 from below. However, as illustrated in FIG. 3, the illumination 23 sandwiches a fourth base plate 11d (not shown in FIG. 3) between which many straight pipes 12 to be inspected are vertically arranged. The camera 21 and the illumination 23 are provided facing each other in the vertical direction so that the camera 21 described later can take the inspection light that has passed through the straight pipe 12.

  Accordingly, the inspection light emitted from the illumination 23 provided below is inspection light that has passed through each straight pipe 12 that is erected perpendicularly to the fourth base plate 11 d, and this inspection light is optically transmitted by the magnifying lens 22. The enlarged image is photographed from the front by the camera 21 provided above. In this way, if the inspection light that has passed through each straight pipe 12 is imaged by the camera 21 and the obtained imaging data is subjected to image processing, if foreign matter remains in the straight pipe 12, the inspection light is blocked. Since the amount of light decreases, the presence or absence of foreign matter in the straight pipe 12 can be automatically inspected by observing a change in the amount of inspection light.

  However, as already described repeatedly, a large number of straight pipes 12 are provided in the spinneret 1. Therefore, as a matter of course, it is necessary to perform an inspection on all straight pipes 12 set up vertically. However, the remaining foreign matter in all straight pipes 12 cannot be inspected only by the combination of the camera 21 and the illumination 23 provided opposite thereto. This is because the camera 21 provided with the magnifying lens 22 limits the range in which an image can be captured, and thus can only capture an image that covers a limited area. As a matter of course, the magnification of the image by the magnifying lens 22 should be a value that can identify the foreign matter attached to each straight pipe 12.

  Therefore, in the spinneret abnormality inspection apparatus according to the present invention (in the following description, it may be simply referred to as “inspection apparatus”), an image is formed in the entire region of the spinneret 1 provided with the straight pipe 12. A mechanism for moving the spinneret 1 to the visual field range of the camera 21 is provided so that data can be obtained.

  That is, the apparatus for carrying out the inspection method of the present invention is the spinneret 1 (substantially the “fourth cap plate 11d”. However, in the following description, these are not distinguished from each other and are simply referred to as “spinner 1”. A hollow type rotary actuator 25 provided with a base for mounting the base is placed, and the spinneret 1 can be rotated. Further, an illumination 23 is disposed below the hollow rotary actuator 25, and the inspection light is irradiated from the illumination 23 toward the spinneret 1 placed in the hollow portion of the hollow rotary actuator 25. .

  The reason why the hollow rotary actuator 25 is used is that each structural unit for spinning one sea-island type composite filament is arranged at equal pitch intervals on multiple concentric circles centering on the die center. It is. As a result, if the spinneret 1 is rotated by a predetermined rotation angle only by determining the position of the spinneret 1 to be inspected so that the center of the spinneret 1 coincides with the center of the hollow rotary actuator 25, Thus, the spinneret 1 can be easily and accurately positioned at the position where the straight pipe 12 that requires this inspection is provided.

  Here, if just a supplementary explanation is given just in case, as is clear from FIGS. 1 and 2, the central axis of each straight pipe provided on the fourth base plate 11d and the fifth base plate 11e. It is necessary to position and install accurately so that the central axis of each polymer merging hole 17 drilled in is substantially coincident with the central axis. For this reason, the fourth cap plate 11d and the fifth cap plate 11e are provided with positioning means such as positioning pins and positioning holes so that they can be positioned accurately during assembly.

  Therefore, by utilizing such a relationship skillfully, positioning pins and positioning similar to the positioning means provided between the fourth base plate 11d and the fifth base plate 11e, in which a large number of straight pipes 12 are erected vertically. A hole is also provided between the fourth cap plate 11d and the “spinner cap abnormality inspection device”. Therefore, the fourth base plate 11d in which a large number of straight pipes 12 are erected vertically can be accurately positioned and installed with respect to the “spinning base abnormality inspection device”.

  Further, in the embodiment of the present invention, as illustrated in FIG. 4, a linear actuator 26 that is slidable linearly in one axial direction with respect to the hollow rotary actuator 25 is attached. As a result, the rotary actuator 25 attached to the hollow base for mounting the spinneret 1 is driven by the linear actuator 26 so as to be slidable in the linear axis direction and the rotary axis direction. It is the composition which becomes. For this reason, the spinneret 1 is accurately positioned at an arbitrary position in a two-dimensional plane along the scanning direction of image capturing by the camera 21 by actuators 25 and 26 that are slidable independently in the linear direction and the rotational direction. It can be moved.

  At this time, drive control of the actuators 25 and 26 is performed by a computer device 27 connected to a drive means including servo motors for driving the actuators 25 and 26. That is, in order to obtain image data required for inspection by giving a number of driving pulses corresponding to the required slide amount to each servo motor that drives each actuator 25 and 26 according to a command from this computer device 27. The actuators 25 and 26 are controlled so as to be slidable in the rotational direction and the linear direction, respectively, by the necessary amount. In addition, this simultaneously controls the spinneret 1 to move to a target position at an arbitrary control speed.

  At this time, although the details will be described later, the computer device 27 also has a function as a spinneret abnormality inspection device, acquires image data photographed by the camera 21, performs image processing, and performs a number of straight pipes. Image processing for detecting whether or not a foreign substance remains in the interior of image 12, and image processing for detecting whether or not a specific straight pipe is deformed in a group of straight pipes 12 to be inspected It can be performed.

  In order to perform the image processing, the computer device 27 has a built-in program for enabling image processing, which will be described later, necessary for image processing, and a function for executing the program, such as a central processing unit (CPU). An image processing program is executed in cooperation with the main device and peripheral devices attached thereto. Note that the computer device 27 stores an interface device as a peripheral device necessary for executing the image processing program to perform image processing, and a memory for storing the image data group captured via the interface device. Needless to say, the apparatus includes various means necessary for processing such as input / output means for inputting / outputting necessary data.

  The camera 21, the actuators 25 and 26, the lights 23 and 24, and the like are attached to the support base 28 as illustrated in FIG. At this time, the illumination 23 attached to the support base 28 will be further described. This illumination 23 has a long life, hardly generates heat, and is not easily scattered when passing through the straight tube 12. It is necessary to irradiate the entire imaging range. From this point of view, the illumination 23 uniformly irradiates a wide area with high brightness and is parallel to the camera 21 and the magnifying lens 22 from the illumination 23 (in the following description, simply abbreviated as “parallel light”). It is preferable to use LED condensing illumination that can irradiate.

  This is because, by using such LED condensing illumination 23, for example, even in the case of a straight pipe 12 having an inner diameter of 1 mm or less, a planar shape from the illumination 23 toward the camera 21 with the spinneret 1 interposed therebetween. This is because the inspection light easily passes through the internal polymer flow path of all the straight pipes 12 in the irradiated region, and the straightness is also improved. In addition, light scattering in the straight pipe 12 is small, and it is possible to perform pseudo three-dimensional imaging inside the straight pipe 12. The illumination 24 attached to the camera fixing column 29 is also preferably an LED illumination that has a long lifetime, is unlikely to generate heat, and can irradiate a wide area uniformly with high brightness.

  In this example, a single-focus optical magnifying lens 22 is mounted as the camera 21 that captures the transmitted light from the LED condensing illumination 23, and noise reduction can be expected. : Pixel), a so-called “megapixel” camera 21 was used.

  As described above, when the image data necessary for the inspection is obtained by the camera 21, for example, the methods described in the above-described Patent Document 4 (particularly, [Area determination processing] and [Perimeter determination processing]) are performed. By using this image data for image analysis, it is possible to detect the presence or absence of foreign matter remaining in each straight pipe 12.

  However, the remaining foreign matter in the straight pipe 12 adheres to any position along the length direction (depth direction) of the straight pipe 12. That is, foreign matter may adhere to a position immediately after the polymer flows into the straight pipe 12, or may adhere to a position immediately before the polymer flows out of the straight pipe 12. In consideration of such points, it is preferable to use the image data photographed with the camera 21 as the image data photographed with focusing at a plurality of locations along the length direction (depth direction) of the straight pipe 12. For this purpose, it is desirable to provide a function capable of adjusting the focal length of the camera 21 at a plurality of locations, and the operation of the camera 21 with respect to the entire surface of the spinneret 1 is desirably performed every time the focal length of the camera 21 is changed.

  The image data photographed by the camera 21 is not limited to image data obtained by photographing only the individual straight pipes 12, and the inspection light passing through a predetermined number of straight pipes 12 is photographed collectively. It is preferable to target image data. This is because image processing efficiency is improved by creating composite image data obtained by synthesizing individual images obtained by capturing the inspection light passing through the individual straight pipes 12 and processing the composite image data at a time. It is because it can be made.

  That is, rather than the inspection by the sequential image processing of the individual image data obtained by photographing the inspection light that has passed through the individual straight pipes 12, the group of straight pipes 12 is subjected to image analysis processing, and the group of straight pipes 12 is subjected to the image analysis processing. From the viewpoint of improving the processing speed, it is preferable to inspect residual foreign matters in the straight pipe at a time using the processing described in Patent Document 4.

  Further, an installation region of a group of straight pipes 12 provided in a group in the spinneret 1 is divided into a plurality of regions, and a plurality of image groups are photographed by operating the camera 21 a plurality of times for these regions. Thus, it is desirable to obtain composite image data necessary for simultaneously inspecting a plurality of straight pipes 12 for abnormalities such as foreign matters remaining in the straight pipes 12 and pipe bending. It is further preferable that the composite image data obtained in this way is subjected to image processing at a time to perform an abnormality inspection process for the spinneret 1.

  In order to achieve the goal of obtaining image data that can be processed as described above, in the present invention, the spinneret 1 is moved to a predetermined plurality of positions by two actuators 25 and 26 that are optimally controlled by the computer device 27. Therefore, it is required to move the positioning with high accuracy. Therefore, the movement amounts (sliding amounts) of the actuators 25 and 26 are input in advance to the computer device 27 as data relating to the design dimensions of the installation area of the straight pipe 12. A program for calculating the amount of movement required for the actuators 25 and 26 based on the inputted design dimension data is stored in the computer device 27.

  By doing so, based on the data information stored in the computer device 27 (the amount of movement of the actuators 25 and 26, etc.), each servo motor is driven via a command from the computer device 27, and the actuators 25 and 26 are driven. Can be accurately positioned. As a result, a predetermined number of image data can be obtained at a predetermined plurality of positions, and composite image data can be created based on the predetermined number of image data.

  In this way, when an image data group obtained by photographing inspection light that has passed through a large number of straight pipes 12 to be inspected is obtained by the camera 21, these image data groups are used as composite image data as described above, and an image processing apparatus. The processing to be taken into the computer device 27 that also serves as the computer is executed. Then, image analysis processing is performed on the composite image data, thereby determining the existence of pipe bending of a large number of straight pipes 12 provided in the spinneret 1 by image processing.

  In the present invention, composite image data is not always necessary, but it is preferable to use composite image data from the viewpoint of obtaining fine image data. . This is because, as will be described later, it is a great feature that pipe bending is detected by the mutual positional relationship (center distance between the straight pipes 12) of the straight pipes 12 provided in the spinneret 1. is there.

  However, this is not the case when inspecting abnormally bent pipes. In other words, in such a case, “(1) the whole group of straight pipes 12 to be inspected for pipe bending is clearly captured, and (2) the polymer pipes of each straight pipe 12 pass through the through holes. If the three requirements are satisfied, it is an image having a resolution that can clearly identify the inspected inspection light, and (3) the distance between the centers of the adjacent straight pipes 12 can be detected. This is because it is not necessary to use synthesized image data, and individual image data (single image data that is not synthesized) can be used.

Hereinafter, a processing method for detecting pipe bending occurring in a specific pipe in a large number of straight pipes 12 will be described in more detail.
First, the spinneret 1 for inspecting pipe bending is positioned and set at a predetermined position between the camera 21 and the illumination 23, and the inspection light projected from the illumination 23 and simultaneously passing through the plurality of straight pipes 12 is used as image data. A process of importing into a computer device 27 (hereinafter referred to as “image processing device 27”) also serving as an image processing device is performed. Therefore, in order to perform this processing, the spinneret 1 is moved to a required position by the actuators 25 and 26 described above, and is moved to a predetermined position for imaging by the camera 21, and the required number of image data is acquired by the camera 21. Shoot. The focal point of the camera 21 is adjusted to the outlet tip of the straight pipe 12 from which the island component polymer is discharged. Further, the group of straight pipes 12 to be inspected is preferably set as “one constituent unit for spinning one sea-island type composite filament” shown in FIG.

  Next, image data captured by the camera 21 is read into the image processing device 27 for image processing. Then, with respect to all the pixels included in the read image data, a threshold value set corresponding to the number of gradations (for example, 256 gradations) of light and darkness (luminance value) of the captured image is set, for example, 256 gradations. At the time of sorting, binarization processing is performed for separating the brightness of each pixel into “bright” and “dark” with reference to the gradation value obtained using the P tile method (Percentile Method).

  In order to correctly perform the binarization process, it is preferable to execute a filter process such as a differential filter or a γ correction in order to remove noise from the acquired image data or enhance the video. After such processing, it is preferable to perform binarization processing on the image data.

  Here, the above-described P tile method (Percentile Method) will be briefly described. Just as described above, the P tile method is divided into “bright” and “dark” by binarization processing as described above. This is a method of binarizing by designating a percentage (%) of the ratio of the image area allocated to “bright” to the entire image area with respect to the entire image area.

  In actual processing, for example, if the image size is 640 × 480 pixels, the total number of pixels is 307200 pixels. Therefore, if the ratio of the image area allocated to “bright” is 37%, for example, the number of pixels corresponding to 37% is 307200 × 0.37 = 113664 pixels. Therefore, a histogram of gradation values when all pixels to be binarized are classified into light and dark gradation values is acquired. Then, when the frequency of the acquired histogram is added from the higher gradation value (from the lower gradation value when the area on the “dark” side is specified), the gradation value exceeding 113664 pixels for the first time is added. This is a method of using a threshold value.

  In the binarization processing by this P tile method, if the ratio of each image area assigned to “bright” and “dark” is constant in all image areas assigned to “bright” and “dark”, There is an advantage that even if the brightness of an image acquired by photographing changes, it is hardly affected. For example, in general, the brightness gradation value of an image obtained by the camera 21 varies depending on the temperature change of the camera 21 body. Also, the illuminations 23 and 24 tend to deteriorate and become darker when used for a long time. Therefore, the P tile method has an advantage that binarization processing can be stably performed as compared with binarization processing by a fixed threshold method using, for example, an intermediate gradation value of the number of gradations as a threshold value. Yes.

  However, the P tile method is not suitable when there is a large variation in the ratio of each image area that is assigned to “bright” and “dark”. However, such a situation does not occur in the binarization process for identifying the existence area of the straight pipe 12 provided in the spinneret 1. This is because, regarding the straight pipe 12 provided in the spinneret 1, the pipe diameter, the number of installed pipes, the pipe arrangement, etc. are uniquely determined when the spinneret 1 is manufactured as originally designed. . For this reason, the ratio of the pixel group recognized as “bright” by detecting each inspection light passing through the through hole of each straight pipe 12 that is a polymer flow path is, of course, substantially unique. It is because it is decided.

  The pixel group recognized as “bright” in the composite image data or the like after the binarization process (the separation process into “bright” or “dark”) described above emits inspection light into the straight tube 12. As long as there is no foreign matter to be blocked, the shape of the polymer channel can be regarded as a region where the inspection light that has passed through the straight through-hole without being blocked is detected. In addition, the pixel portion determined to be “dark” at this time can be regarded as a portion where the through hole of the straight tube 12 is not present and the inspection light does not reach the camera 21.

  However, since the inspection light that has passed through the straight pipe 12 that has been bent at its distal end must pass through the through-hole of the bent straight pipe 12, it is reflected once on the through-hole wall, and so on. The optical path is bent. For this reason, as compared with the case of passing through a straight through hole, there is a difference in the degree of bending, but there is a shift in the existing region of the pipe tip where the inspection light is detected. As a result, even in the pixel that detects the inspection light and recognizes it as “bright”, there is a difference between the two.

  By analyzing the image of all the pixels recognized as “bright” by the above-described processing, the existence region of the front end portion (polymer outlet) of the straight pipe 12 is first identified, and then the deviation is calculated. By doing so, it is possible to inspect whether or not the pipe is bent. Specifically, in this process, for example, “1” is assigned to all pixels identified as “bright” by detecting the inspection light passing through the through hole of the straight pipe 12 by the P tile method, for example, and “dark”. Binarization is performed by assigning, for example, “0” to all the pixels identified as “”. At this time, the pixel portion identified as “dark” is, as a matter of course, a portion where the through hole of the straight tube 12 through which the inspection light passes does not exist.

  The pixel data related to each straight pipe 12 that has been determined to be “bright” in the binarization process is regarded as the existence area of the straight pipe 12 even though noise is mixed and the existence area of the straight pipe 12 is not present. Occasionally occurs. Therefore, by this binarization processing, ““ bright ”:“ 1 ”” continuously becomes a cluster of a plurality of pixels from the entire pixel data group distributed to ““ bright ”:“ 1 ””. Detect where it exists.

  As described above, if a collection of continuous pixels composed of ““ bright ”:“ 1 ”” is detected, it is not determined that ““ bright ”:“ 1 ”” by noise. It can be regarded as an aggregate of pixel groups that have detected the inspection light that has surely passed through the through hole of the straight pipe 12. Therefore, it is possible to recognize that each “pixel group aggregate (hereinafter referred to as particles)” recognized in this way is derived from the inspection light that has passed through each straight tube 12. Therefore, the center position of the tip polymer flow path of each straight pipe 12 can be detected by calculating the center coordinate position of the tip polymer flow path of each straight pipe 12 where this “particle” exists. Become.

In this way, when the center coordinate position regarding the existence region of the polymer flow path tip of each straight pipe 12 is obtained by image processing, next, as a second process, a specific straight pipe 12 in the group of straight pipes 12 is selected. An identification process is performed to determine whether or not pipe bending has occurred in the pipe 12.
Here, the identification process will be described more specifically. First, as described above, detection of the existence region of each straight pipe 12 from each “particle” data detected as a ““ bright ”:“ 1 ”” lump. Perform processing. This detection process is a process of calculating the center coordinate value of each “particle” portion, that is, the spread of the pixel aggregate consisting of ““ bright ”:“ 1 ”” (the number of pixels constituting each “particle”). is there.

Note that this process assumes that there are n total pixels allocated to ““ bright ”:“ 1 ”” constituting each particle, and the X coordinate of an arbitrary pixel in each “particle”. And Y coordinates are represented by (X _i , Y _i ), the coordinates (X _oi , Y _oi ) of the center position O _i of the tip polymer flow path of an arbitrary “particle (straight pipe 12)” The central coordinates O _i (X _oi , Y _oi ) of the straight pipe 12 are obtained by using the fact that they can be expressed as follows.

  Using the center coordinate value of the tip polymer flow path of each straight pipe 12 obtained by this process, the center coordinate value of the tip polymer flow path of any straight pipe 12 and the straight pipe 12 as a reference for comparison. The distance between the center coordinates between the tip polymer flow path (for example, the straight pipe 12 arranged adjacently) is calculated. When the distance between the center coordinates is obtained in this manner, for example, the distance value between the center coordinates recognized that the straight pipe 12 is bent is, of course, when the pipe is not bent (from the data at the time of design). The distance between the center coordinates of the polymer flow path at the tip of the arbitrary straight pipe 12 can be calculated in advance as a default value), but there is a difference in degree.

  Therefore, if the difference in the distance between the central coordinates is larger or smaller than the allowable value compared with the determination reference value, it can be determined that the pipe bend exists. However, it goes without saying that the discrimination reference value and the allowable value are set to optimum values in advance through experiments or the like.

Another method is described below. That is, it is possible to accurately specify the position of each straight pipe 12 (the position of the central axis) from the drawing data at the time of design and the acceptance inspection data at the time of completion of production. Therefore, the image data taken by the camera 21 is superimposed on the above-described center coordinate O _i (X _oi , Y _oi ) of the polymer flow path at the tip of the straight pipe 12 obtained by image processing, and how much the center position is. The occurrence of pipe bending can also be detected by the computer device 27 (image processing device 27) automatically determining whether or not there is a deviation.

  In this way, when it is determined that a pipe bend exists in a certain straight pipe 12 in the group of straight pipes 12 to be inspected for pipe bend, the pipe bend is accurately associated with the position of the straight pipe 12 at the initial design. Is straightly identified and stored in the computer device 27 (image processing device 27). In some cases, the straight pipe 12 where the pipe is bent is marked.

  Then, the inspector can easily take out information related to “the specific straight pipe 12 in which the pipe bend is detected” stored in the computer device 27 (the image processing device 27), and the computer device 27 (the image processing device 27). The specific straight pipe 12 where the pipe bends can be easily corrected based on the information extracted from (1).

  The data after the presence of pipe bending or residual foreign matter has been inspected is stored in a storage device based on the position information of each straight pipe 12, so that the operator in charge of inspection can visually confirm the data. It can also be. At this time, as a matter of course, the straight pipe 12 in which an abnormality has occurred is obtained by performing an operation of correcting pipe bending or removing residual foreign matter using a jig or the like while using a magnifying glass while visually confirming. Can be returned to a normal state.

  In addition, according to the present invention, it is possible to easily and automatically perform inspection of pipe bending of the straight pipe 12 provided in the spinneret 1 used at the time of synthetic fiber production, and further, inspection of foreign matters in the straight pipe 12. As a result, labor savings can be achieved.

1: Spinneret 2: Abnormality inspection device 11a to 11f: 1st to 6th cap plate 12: Straight pipe 13: Island component polymer introduction hole 14: Sea component polymer introduction hole 15: Island component polymer reservoir 16: Sea component polymer reservoir 17: Polymer junction 18: Funnel-shaped channel 19: Discharge hole 21: Camera 22: Magnifying lens 23: Illumination 24: Illumination 25: Rotary actuator 26: Linear actuator 27: Computer device 28: Support base 29: Camera fixing column

Claims (4)

The following (1) to (3) for detecting at least pipe bending occurring in the pipe in a spinneret in which a large number of pipes having a circular channel cross-sectional shape extending in a straight line are vertically arranged with respect to the base plate. Spinneret abnormality inspection equipment with the requirements of
(1) Illumination for irradiating inspection light that passes through a through-hole that is a polymer flow path of the pipe vertically installed;
(2) A magnifying lens is provided to face the illumination with the base plate in between, and to magnify the inspection light irradiated from the illumination and passed through the group of through holes to a predetermined magnification. Camera
(3) The presence region of the polymer flow path at the pipe tip is specified from the image data obtained by photographing the inspection light that has passed through the through hole with the camera, and the center of the polymer flow path at the tip of the pipe from the specified existence region Calculate the coordinate value, and the calculated center coordinate of the polymer flow path at the tip of each pipe is more than the allowable value than the center coordinate value of the polymer flow path at the tip of each pipe determined at the time of designing the spinneret. An image processing apparatus having an identification function for discriminating that pipe bending occurs when it occurs.   The spinneret abnormality inspection apparatus according to claim 1, wherein the pipe is a pipe for discharging an island component polymer into a sea component polymer when melt-spinning the sea-island type composite fiber. The following (1) to (5) for detecting at least pipe bending occurring in the pipe in a spinneret in which a large number of pipes having a circular channel cross-sectional shape extending in a straight line are erected vertically with respect to the base plate. ) Abnormal inspection method for spinneret with requirements
(1) Irradiate inspection light that passes through a through-hole that is a polymer flow path of the pipe installed vertically,
(2) The inspection light that has passed through the through hole is photographed as image data by enlarging the inspection light to a predetermined magnification at the pipe tip on the side from which the polymer is discharged,
(3) The image data is binarized with a gradation of light and dark with a predetermined threshold as a reference, and the pixel in which the inspection light passing through the through hole is detected is “bright”, and the inspection light is detected. The pixels that were not done are classified as “dark”
(4) Identify the existing region of the aggregate of pixels separated from "bright", calculate the center coordinate value of the polymer flow path at the tip of each pipe from the specified existing region,
(5) The calculated center coordinates of the polymer flow path at the tip of each pipe are uniquely determined at the time of designing the spinneret. If a pipe is bent, it is identified.   The spinneret abnormality inspection method according to claim 3, wherein a threshold value for allocating pixels constituting image data to "bright" or "dark" by the P-tile method in the binarization process is determined. .

Images