JPS63120164A - Method for measuring number of twists - Google Patents

Abstract

(57) [Abstract] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

DETAILED DESCRIPTION OF THE INVENTION (Industrial Application Field) The present invention relates to a twist number measuring method for measuring the twist number of a running or stationary twisted yarn using an industrial television.

(Prior art and its problems) It is well known that in fiber processing, particularly in the false twisting process, the number of twists applied during heating has a great effect on the properties of the yarn obtained after untwisting. For this reason, various attempts have been made to measure the number of twists of twisted yarn while it is running in order to obtain a uniform product. As an example,
There is a method described in In this method, the running or stationary twisted yarn is continuously or pulsed illuminated, and an industrial television is used to take an enlarged image of the yarn at a position where the twisted yarn axis direction and the horizontal scanning line direction are parallel. The number of twists is determined by obtaining an image having bright and dark areas (bright line areas) corresponding to the unevenness of the twist shape, shaping the video signal including the image signal into a pulse signal, and counting the pulse signal during a predetermined horizontal scanning time. This is a method of measuring

However, with this method, if the imaging signal contains noise, the noise may not be removed even after pulse signal shaping. For this reason, there was a possibility that accurate counting would not be performed. In addition, the counting error results in a large head circumference due to the blurring of the image of the thread. For this reason, it is necessary to take measures such as suppressing the fluctuation of the yarn and using an optical system with a large depth of focus, and in addition, the yarn must be twisted at a position where the axial direction of the yarn is parallel to the horizontal scanning direction of the television camera. It is also necessary to take an image,
There was a problem in that it required a considerable amount of technical effort.

(Objective of the Invention) J The object of the present invention is to solve the above-mentioned conventional problems, to be able to accurately measure the number of twists regardless of the angular relationship between the axial direction of the twisted yarn and the horizontal scanning direction of the television camera, and to It is an object of the present invention to provide a method for measuring the number of twists that can be easily configured as necessary depending on the situation so as not to be affected by noise and the contrast of an image of the twisted yarn.

(Means for Achieving the Object) In order to achieve the above object, the twist number measuring method according to the present invention uses an industrial television to illuminate running or stationary twisted yarn continuously or in pulses. Obtaining an image including a bright line portion, extracting bright line portion data from the image,
From the bright line part data, find the average pitch between the bright line parts,
The number of twists of the twisted yarn is determined based on this pitch.

(Example) A, j and FIG. 1 are schematic explanatory diagrams of a twist number measuring device used in an example of the present invention. In the figure, an illumination device is used for the yarn 1 whose twist number is measured! 2 and an industrial television camera 3 are adjusted and installed at appropriate positions.

Lighting equipment! ! 2 can be illuminated not only with continuous light but also with pulsed light using a strobe flash, and when measuring the number of twists online, the latter type of illumination is usually used. The yarn 1 is enlarged and imaged by a television camera 3, and this image is displayed on a monitor device 4. The operator can adjust the television camera 3 based on the thread *5 displayed on the monitor @4.

Further, an image signal 7 containing image information of the thread 1 obtained by the television camera 3 is taken into the video memory 6. The video memory 6 digitally converts the image signal, for example, 25
It is stored in the built-in memory as luminance data of approximately 6x256 (or 512x512) pixels. The number of twists of the brightness data stored in the video memory 6 is calculated by the image processing device 7 according to the method described in detail below.

Note that the yarn applicable to this invention includes nylon,
It includes not only synthetic fibers such as PET (polyethylene terephthalate) but also gold twisted wires and the like.

B. Twist calculation formula Once the twist pitch p (unit: number of pixels) of yarn 1 is determined from the brightness data stored in the video memory 6, the number of twists of yarn 1 can be determined from the following equation.

N=1000/(p-k-f)...(1) However, N: number of twists (Tn1): pixel pitch (rtm/pixel), and this value is based on the magnification rate of the TV camera 3, etc. This is the pitch corresponding to the actual effective field of view.

f: This is the number of single yarns that emerge from the outer periphery in one twist of yarn 1, and can be determined experimentally.

00. FIG. 2 is a block diagram showing the flow of image processing in the image processing device 7. As shown in FIG. This will be explained below with reference to FIG.

(The luminance data taken into the C-11 binarized video memory 6 is binarized using an appropriate threshold value. The threshold value at this time is determined by a method based on a calculation formula according to the luminance distribution of the luminance data [automatic threshold value, By setting the value to 1, the influence of the brightness of the illumination device 2 on the binarization can be avoided as much as possible.
The converted pixel data is Vl.

(C-2) Extract the bright line parts from the binarized pixel data ■1 by labeling (C-1), and define each bright line part as bl, b2. It is recognized as independent data such as b3... and numbered as shown in 2.

(C-3) Each bright line portion b1 labeled by gravity center calculation (C-2). b2
,... Find the center of gravity as shown in ■3. Let these barycenter coordinates be (Xo, y,) (i is the bright line subscript number). At this time, it is desirable to rearrange the items in ascending order of X.

(C-4) From the barycenter coordinates (×・, Vi) obtained by uniform pitch (C-3),
The average distance pi between the centers of gravity of each bright line portion is determined. Here, when determining the distance p between the centers of gravity of each bright line part, the operator can
It can be divided into several cases by observing.

(2) When the yarn 1 is approximately parallel to the horizontal scanning line, the Y coordinate of the center of gravity is approximately constant, so the distance p1 between the centers of gravity of each bright line portion can be approximated by the following equation.

pi ``xi+1-xi ``””)■
Yarn 1 has an inclination with the horizontal scanning line on the screen, but the center of gravity of each bright line is almost in a straight line... (3) ■ Yarn 1 has an inclination with the horizontal scanning line on the screen. If the center of gravity of each bright line portion is not on a straight line, the center of gravity of each bright line portion is projected onto a straight line L1 parallel to yarn 1 at each coordinate (X, /) as shown in FIG.

After finding y,'), the distance between the centers of gravity p...
(4) Note that the straight line L1 parallel to the yarn 1 is determined by a regression line fitted by least squares approximation using the barycentric coordinates of each bright line portion.

Furthermore, when the shape of each bright line part 2 is observed to be elongated in an elliptical shape as shown in FIG. Coordinate(
x・, Vi), and the coordinates of the intersection of thread 1 and parallel straight line L1 are (X+').

y,′), find the distance between the centers of gravity p1 from equation (4). It is more accurate to

The average pitch is determined by the following equation using each distance pi between the centers of gravity obtained by any of the above methods.

p=(Σp・ )/(n-1> ...(
5) λ・Ll n: Number of bright line portions Therefore, even if yarn 1 has an inclination with the horizontal scanning line on the screen, by applying the above method The distance between the centers of gravity can be calculated.

(C-S) By substituting the average pitch p obtained in the twist number calculation (C-4) into equation (1), the number of twists N (tems/m) is found.

D. The image captured by the abnormal value detection TV camera 3 is unclear,
If the threshold value for converting to 21ia in (C-1) is inappropriate, bright line omissions 8 and noise 9 may be inserted into the image data taken into the video memory 6, as shown in FIG. In this case, the abnormal data is detected and removed by the method described below, and then the average pitch calculation (C-4) is performed.

■ After calculating the average pitch p in (C-4), if the distance pi between the centers of gravity of each bright line portion is, for example, pi ≧ 1.5p or i ≦ 0.5p, then the distance pi between the centers of gravity is considered to be abnormal data. It is assumed, removed, and the average pitch P is calculated again.

(2) Bright line parts whose center of gravity coordinates are separated by a predetermined distance or more from the regression line L1 determined from the center of gravity of each bright line part are regarded as abnormal data and are removed.

(2) When labeling in (C-2), the area of each bright line portion is determined, and those that do not reach a predetermined value are considered to be noise and removed.

By adopting one or more of the jδ solid abnormality detection methods described in ■ to ■ above, even if bright line parts are missing or noise is inserted due to blurring of the image of yarn 1 as shown in Fig. 5, An accurate average pitch p can be determined.

(Effects of the Invention) As described above, according to the present invention, it is possible to accurately measure the number of twists of the twisted yarn regardless of the angular relationship between the twisted yarn axis direction and the horizontal scanning direction of the monitor device, and the imaging situation If necessary, it is possible to easily configure the image data so as not to be affected by out-of-focus effects caused by noise inserted into image data and images with weak contrast.

In addition, it can measure both static and running yarns, and the number of twists can be measured without contacting the yarn. Furthermore, if you use short-term pulse lighting devices such as strobes and flashes,
It is possible to measure the number of twists of yarn running at high speed, and by using a monitor, it is easy to adjust the image, and of course it is possible to constantly monitor the measurement status.

[Brief explanation of the drawing]

Fig. 1 is a schematic explanatory diagram of a twist number training device used in an embodiment of the present invention, Fig. 2 is a block diagram showing the processing flow of the image processing device shown in Fig. 1, and Fig. 3 is a center of gravity of the bright line portion. A diagram showing the projection of the coordinates onto a straight line parallel to the thread. Figure 4 is a diagram showing the same projection when the bright line part in Figure 3 is an ellipse. Figure 5 is a diagram showing the projection of the coordinates onto a straight line parallel to the thread. Figure 5 shows noise, blur, etc. inserted into the bright line data. This is a diagram showing the 1... Yarn, 2... Lighting device, 3...
TV camera, 4...Monitor device, 6...Video memory, 7...Image processing device Fig. 1 Fig. 2

Claims (7)

[Claims] (1) Using an industrial television to illuminate running or stationary twisted yarn continuously or in pulses, obtain a yarn image including a bright line portion, extract bright line portion data from the yarn image, and extract the bright line portion data from the yarn image. A method for measuring the number of twists, characterized in that the average pitch between the bright line parts is determined from part data, and the number of twists of the twisted yarn is determined based on this pitch. (2) The twist number measuring method according to claim 1, wherein the average pitch between the bright line portions is determined by determining the center of gravity of the bright line portion data and is the average distance between the centers of gravity. (3) The average pitch between the bright line portions is determined by determining the center of gravity of the bright line portion data, projecting the center of gravity onto a straight line parallel to the axial direction of the twisted yarn, and obtaining the average pitch between the data after projection. The method for measuring the number of twists according to item 1. (4) The average pitch between the bright line parts is the average pitch after abnormal data is removed from each bright line part data by the abnormal data discrimination means.
The method for measuring the number of twists described in any of paragraphs. (5) The twist number measuring method according to claim 4, wherein the abnormal data determining means determines whether the distance between the bright line portion data is within a predetermined range. (6) The number of twists according to claim 4, wherein the abnormal data determining means determines whether the distance between the bright line portion data and the regression line determined from the bright line portion data is within a predetermined range. Measuring method. (7) The twist number measuring method according to claim 4, wherein the abnormal data determining means determines whether the area of the bright line portion data is equal to or larger than a predetermined size.