JPH0160092B2 - - Google Patents

Description

[Detailed Description of the Invention] Industrial Application Field The present invention is directed to the treatment of flaws caused by the spacing between stitches of knitted fabric being wider than the regular spacing between stitches, so-called flaws called needle streaks, and defects where stitches are missing. This invention relates to a device for detecting defects such as hole-type defects.

2. Description of the Related Art In recent years, detection devices using photodetectors such as phototransistors and television cameras have come to be used as means for detecting knitting defects produced in knitted fabrics knitted by knitting machines. This detection device generally uses one detection element to detect the amount of transmitted light from the fabric, compares it with a certain threshold value, and detects anything exceeding this threshold value as a flaw signal. However, this device has the disadvantage that it cannot accurately detect flaw signals due to fluctuations in the background and light intensity level of the light source due to the influence of external light, etc.
To solve this problem, a device has been developed that calculates the average value of the first-order delay element of the signal, calculates the difference between this average value and the actual measurement value, and detects defects by determining whether this difference exceeds a predetermined threshold. .

Problems to be Solved by the Invention By the way, the above-mentioned needle streak is a defect in which the thread spacing is about 50% wider than in the regular stitch, and is generally caused by mechanical problems such as bent needles. Occurs due to trouble. Therefore, if needle streaks occur, it is necessary to immediately stop the operation of the knitting machine.

However, needle marks have smaller fluctuations in the light level than hole-based knitting defects, so if the method of sorting using a fixed threshold value is used, the needle marks will be judged as noise due to external light, etc., and will not be detected as a flaw, and the needle marks will not be detected as a flaw. Since the rise of the signal is not very steep, if the average value of the signal is taken, there is a risk that the difference will not exceed a certain value and will not be detected as a flaw because the average value changes following the actual measurement value.

As described above, the conventional detection device as described above cannot detect needle lines as knitting defects, which is a fatal drawback in promoting automation of knitting machines.

Means for Solving the Problems The present invention has been made in view of the above-mentioned circumstances, and is capable of reliably detecting needle lines and also detecting hole-based knitting flaws. The object of the present invention is to provide a detection device for defects.

The present invention provides an apparatus for optically detecting flaws in a knitted fabric where regular stitches are not formed, which uses two or more light beams arranged at a pitch that is an integral multiple of the spacing between yarns forming the stitches or at a pitch close to this. a detector;
An arithmetic unit that adds and subtracts the output signals from each photodetector and a certain detection judgment value based on the presence or absence of a flaw are set, and a comparison of the magnitude of this detection judgment value and the output signal from the arithmetic unit detects a flaw. The configuration includes a discriminator for discriminating the presence or absence of the section.

Effects According to the present invention, when a flaw appears or appears within the detection range of each photodetector, the calculation result of the arithmetic unit changes, and the discriminator determines the difference between the calculation result and the detection discrimination value. The presence or absence of a flaw is determined based on the comparison result.

Embodiments Hereinafter, embodiments of the present invention will be described with reference to the drawings.

FIG. 1 is a block diagram showing a knitting flaw detection device according to the present invention, and this embodiment describes a case in which the device is attached to a circular knitting machine to detect knitting flaws. In the figure, 1 is a detection camera, 2 is a light source, and 3 is a detection box. Further, X is a knitted fabric which is knitted into a cylindrical shape in a circular knitting machine (not shown) and sent downward, and this knitted fabric X has stitches (flat knitting) shown in FIG. 16. The downward feeding of the knitted fabric X is performed so that the knitted fabric X moves by a predetermined stroke every time the knitted fabric X makes one rotation in the circumferential direction (in the direction of the arrow Y in the figure). That is, the detection camera 1 is installed on the outer circumference side of the knitted fabric X, and while the knitted fabric X rotates, the light emitted from the light source 2 installed inside the knitted fabric X passes through the knitted fabric X. The amount is continuously collected as a light and dark signal.

As shown in FIG. 2, the detection camera 1 includes two vertically elongated photodetectors (hereinafter simply referred to as detectors) 4a and 4b made of photoelectric conversion elements such as silicon solar cells.
Built-in. These detectors 4a, 4b are arranged at a pitch twice as large as the interval between the warp threads forming the stitches of the knitted fabric X.
4b is such that, if the warp threads forming the stitches are at regular intervals, different warp threads located on both sides of one warp thread can be brought into the detection range in the same state. The bright and dark signals obtained by the detectors 4a and 4b are given to a subtraction section 6a and an addition section 6b.

Here, if the level of the bright/dark signal obtained by the detector 4a is A, and the level of the bright/dark signal obtained by the detector 4b is B, then the flaw signal level Z ₁ output by the subtractor 6a is Z ₁ =B-A ...obtained as (a). On the other hand, the adder 6b calculates the sum of the bright and dark signals, and the flaw signal level Z ₂ output from the adder 6b is obtained as Z ₂ =A+B (b).

The flaw signal levels Z ₁ and Z ₂ are output to a detection box 3, respectively, and the detection box 3 includes discriminating circuits 7a and 7b for discriminating whether the input flaw signal is higher than the detection level, and discriminating circuits 7a and 7b for discriminating whether the input flaw signal is higher than the detection level or not. A built-in alarm circuit 8a and a counting circuit 8b are built in to perform respective operations based on the outputs of the circuits 7a and 7b.

When the light and dark signals of the knitted fabric are collected by the detectors 4a and 4b, which are arranged at twice the pitch of the warp threads forming the stitches as described above, the warp threads are arranged at a pitch that is twice the pitch of the warp threads forming the regular stitches. If they are lined up, each of the detectors 4a and 4b obtains bright and dark signals in the same state regardless of the movement of the knitted fabric X, as shown in FIGS. 3A to 3C. (The portion indicated by the dashed line in the figure indicates the detection range of each detector 4a, 4b.) That is, in the state shown in FIG. ) The light and dark signals are obtained when W1 and W3 are detected, and as the knitted fabric Then, as shown in FIG. 3C, a light/dark signal is obtained by detecting the warp yarns W1 and W3 at the same position on the right side of the detection range. Therefore,
When the warp yarns of the knitted fabric are at regular intervals, each detector 4a, 4b always obtains the same brightness signal, and therefore the value of the above equation (a) is always "0".

Next, a case in which the detectors 4a and 4b detect a needle thread whose distance between warp yarns is farther apart than in the case where the interval is normal will be described with reference to FIGS. 4A to 4E. In the figure, the regular stitch spacing is between warp yarns W11, W12, and W13 and between warp yarns W14, W15, and W16, but there is a needle thread H between warp yarns W13 and W14.
The warp spacing is about 50% wider than in the regular part. As shown in FIG.
While b is getting bright and dark signals with the warp threads W14 and W16 within the detection range, each detector 4a and 4b is getting bright and dark signals in the same state, and the value of the above formula (a) is "0". be. However, as the knitted fabric moves, as shown in FIG. 4B, when the warp W14 passes through the detection range of the detector 4a, other detectors 4
Although the warp yarn W15 is included in the detection range in b, there is a needle line H in the detection range of the detector 4a, and the warp yarn W13 is not included. Therefore, the amount of light transmitted by the light source 2 in the detection range of the detector 4a increases, and the level of the bright/dark signal detected by the detector 4a becomes higher than the level of the bright/dark signal detected by the detector 4b. In this way, when the level of the bright/dark signal detected by the detector 4a becomes higher than the level of the bright/dark signal detected by the detector 4b, the above equation (a) has a negative element.

As the knitted fabric further moves, the detectors 4a and 4b detect warp yarns W13 and W, respectively, as shown in FIG. 4C.
14 within the detection range to obtain a bright/dark signal, the value of equation (a) changes from negative to positive in the vicinity of O.

As the knitted fabric moves further and the warp yarn W14 passes through the detection range of the detector 4b as shown in FIG. There is a needle line H in the range, and the warp thread W13 does not enter the detection range. Therefore, equation (a) has a positive element.

Then, as shown in FIG. 4E, the detectors 4a,
4b comes to include warp threads W11 and W13 respectively within the detection range, the value of the above formula (a) becomes "0" again.
becomes.

The output of the subtraction unit 6a that changes as described above is expressed in a graph as shown in FIG. _t1 to _t5
represent the times in the states shown in FIGS. 4A to 4E, respectively. The discrimination circuit 7a is
In FIG. 5, when the output of the subtractor 6a exceeds the detection level _Z10 , the flaw detection signal is sent to the alarm circuit 8a.
When the alarm circuit 8a receives this flaw detection signal, it outputs a stop signal to stop the operation of the circular knitting machine from the output terminal ALM, and also generates an alarm sound using a buzzer 9a or the like.

Next, if a round hole exists in the knitted fabric X, it is detected as follows. As shown in Figures 3A to 3C, when both the detectors 4a and 4b have the normal stitch portion within the detection range, the value of equation (b) above is within a substantially constant range. . However, if there is a round hole F as shown in FIG. 6, which is much larger than the needle thread described above, the value of the above equation (b) becomes large depending on the amount of transmitted light. Figure 7 is a graph of the value of equation (b) above when a round hole is detected in this way, that is, the output of the adding section 6b. This shows the case where a round hole is being detected. _Z20 indicates a detection level, and when the output of the adder 6b exceeds this detection level _Z20 , the discrimination circuit 7b outputs a flaw detection signal to the counting circuit 8b, and this counting circuit 8b receives the signal. The number of detected round holes is displayed on a display section provided on the side of the detection box 3.
The counting by the counting circuit 8b is reset by inputting a reset signal generated by the circular knitting machine from knitting the required length of cloth from the particle RST.

In the above embodiment, two detectors are used to detect the flaws, but the present invention is not limited to this, and it goes without saying that three or more detectors may be used to detect the flaws. It is.

FIG. 8 is a block diagram showing another flaw detection device according to the present invention, which uses three detectors to detect flaws. In the figure, 4a, 4b, and 4c are detectors, respectively, and if the warp threads forming the stitches are at regular intervals, they are arranged at a pitch twice as large as the regular interval. 5 is a weighting section, 51 is an amplifier built in the weighting section 5, 61a is an addition/subtraction section, 62a, 63a
are an adder and a subtracter built into the addition/subtraction section 61a, and the other components are the same as in FIG. 2.

The bright and dark signals obtained by the detectors 4a, 4b, and 4c are sent to the weighting section 5, and the bright and dark signals obtained by the detector 4b are amplified twice in the amplifier 51. That is, in the weighting section 5, "1" is assigned to the bright and dark signals obtained by the detectors 4a and 4c.
However, the bright and dark signals obtained by the detector 4b are each weighted by "2". and,
Each of the weighted bright and dark signals is sent to the adder/subtracter 61a, and the added value obtained by the adder/subtracter 61a is output to the detection box 3. In the adder/subtractor 61a, an adder 62a first adds the brightness signals obtained by the detectors 4a and 4c, and then a subtracter 63a adds the brightness signals obtained by the detector 4b, which have been amplified twice by the amplifier 51. The output value of the adder 62a is subtracted from the bright/dark signal. That is, if the level of the bright/dark signal obtained by the detector 4a is A, the level of the bright/dark signal obtained by the detector 4b is B, and the level of the bright/dark signal obtained by the detector 4c is C, then the addition/subtraction unit 61a outputs The flaw signal level Z ₃ is obtained as Z ₃ =2×B−(A+C)=(B−A)+(B−C) (c).

On the other hand, the brightness and darkness signals obtained by the detectors 4a, 4b, and 4c are processed by an addition section 6 configured separately from the addition/subtraction section 61a.
The adder 6b calculates the sum of the bright and dark signals, and this sum is also output to the detection box 3. That is, the addition section 6
The flaw signal level Z ₄ outputted by b is obtained as Z ₄ =A+B+C (d).

When the light and dark signals of the knitted fabric are collected by the detectors 4a, 4b, and 4c arranged corresponding to the spacing of the warp yarns forming the stitches as described above, it is found that the spacing of the warp yarns is at the predetermined spacing forming the regular stitches. If they are lined up, each of the detectors 4a, 4b, 4c obtains bright and dark signals in the same state regardless of the movement of the knitted fabric X as before. Therefore, the value of the above equation (c) is always "0".

Next, a case in which the detectors 4a, 4b, and 4c bring into the detection range a needle thread in which the warp threads are spaced further apart than in the normal case will be described with reference to FIGS. 9A to 9E. In the figure, warp threads W9, W10, W1
1, W12, W13 and warp threads W14, W15,
The stitch spacing between W16, W17, and W18 is normal, but there is a needle line H between the warp yarns W13 and W14, and the warp spacing is about 50% wider than in the normal portion. As shown in FIG. 9A, the detectors 4a, 4b,
Each detector 4a,
4b and 4c have bright and dark signals in the same state, and the value of the above equation (c) is "0". However, as the knitted fabric moves, as shown in FIG. 9B, when the warp yarn W14 passes through the detection range of the detector 4a,
The other detectors 4b and 4c are warp yarns W15 and W, respectively.
17 is included in the detection range, there is a needle line H in the detection range of the detector 4a, and the warp thread W13 does not enter the detection range. Therefore, the amount of light transmitted by the light source 2 in the carry-out range of the detector 4a increases, and the level of the bright/dark signal detected by this detector 4a becomes higher than the level of the bright/dark signals detected by the other detectors 4b, 4c. In this way, when the level of the brightness signal detected by the detector 4a becomes higher than the level of the brightness signal detected by the other detectors 4b and 4c, the above equation (c) becomes B
Since -C is "0", it has a negative element equivalent to B-A. In this way, the value BA of the above equation (c) when only the detector 4a does not include any warp threads within the detection range is temporarily set to "-M" (provided that M>0).

As the knitted fabric moves further and the needle H enters the detection range of the detector 4b as shown in FIG. The level of the bright/dark signal detected by the detector 4b becomes higher than the level of the bright/dark signal detected by the other detectors 4a, 4c. Detector 4b
Since the output of is weighted by "2" in the weighting section 5, the above equation (c) becomes one with a positive element as the value of B increases. As mentioned above, if the value of equation (c) is set to "-M" when only the detector 4a does not put any warp threads within the detection range, then only the detector 4b puts any warp threads into the detection range. Since B-A and B-C are each "+M" when not included, the value of equation (c) at this time is "+2M".

When the knitted fabric moves further and the detector 4c brings the needle thread H into the detection range as shown in FIG. 9D, the detectors 4a and 4b detect the warp threads W10 and W, respectively.
12 is included in the detection range, the above equation (c) has a negative element as the value of C increases, similar to when the detector 4a described above puts the needle H in the detection range. , for example, has a value of "-M". Then, as shown in FIG. 9E, the detector 4
a, 4b, 4c are warp threads W9, W11, W, respectively
13 comes within the detection range, the value of the above equation (c) becomes "0" again.

The output of the addition/subtraction section 6a changing as described above is expressed in a graph as shown in FIG. 10. t _1
-t5 represent the times in the states shown in FIGS. 9A to 9E, respectively. The discrimination circuit 7
In FIG. 10, when the output of the adder/subtractor 61a exceeds the detection level _Z30 , the alarm circuit 8a issues a flaw detection signal to the alarm circuit 8a. It outputs a stop signal to stop the operation of the knitting machine, and also generates an alarm sound using a buzzer 9a or the like.

Next, if a round hole exists in the knitted fabric X, it is detected as follows. Round hole F as shown in Figure 11
If , the value of the above equation (d) increases depending on the amount of transmitted light. Figure 12 is a graph of the value of equation (d) above when a round hole is detected in this way, that is, the output of the adding section 6b. shows a case where a round hole is detected. _Z40 indicates a detection level, and when the output of the adder 6b exceeds this detection level _Z40 , the discrimination circuit 7b outputs a flaw detection signal to the counting circuit 8b, and the counting circuit 8b receives the signal. The number of round holes detected is counted and displayed on the display section provided on the side of the detection box 3.

In the embodiment shown in FIG. 8, the weighting applied to the signal obtained by the detector can be arbitrarily changed in accordance with the subsequent signal processing process. Further, the method of addition or addition/subtraction of brightness and darkness signals is not limited to the above embodiment. Furthermore, although the addition/subtraction section 61a in the above embodiment is composed of an adder 62a and a subtracter 63a, the subtracter 63a
Alternatively, a differential amplifier may be used, and the output thereof may directly drive the alarm circuit. Furthermore, although the above embodiment is divided into a camera and a detection box for convenience, it is of course possible to integrate them into one unit.

FIG. 13 is a plan view showing the positional relationship between the optical axis A of the light emitted from the light source 2 and the knitted fabric X. FIG. In this figure, if the angle between the knitted fabric X and the optical axis A is α, it is preferable to adjust this α so that 40°≦α≦90°. This is because if α is smaller than 40°, even if a flaw falls within the detection range of the detector, the projected area of the thread will be larger than when α=90°, and the sensitivity will decrease.

The flaw detection sensitivity of the detector improves as the length in the warp direction becomes longer, but if the length is too long, precision is required for installation adjustment. Therefore, the length of the detector is
Knitting length that the circular knitting machine knits in one rotation, e.g. 0.5 to 100
A range of 1 to 5 times mm is preferable.

FIG. 14 is a front view showing an example of the configuration of a detector capable of improving flaw detection sensitivity, and corresponds to FIG. 8 in which three detectors are provided. That is, three detectors are divided into two in the longitudinal direction of each detector to form two sets (six detectors), with a separation of 200 μm or less between them,
As shown in FIG. 15, each divided set of detectors 4
a, 4b, 4c and 41a, 41b, 41c are connected to weighting sections 5, 5', respectively, and the output sides of addition/subtraction sections 61a, 61a' to which these output signals are given are either left as they are or connected to a rectifier circuit (not shown). is connected to the adder circuit 64 via the adder circuit 64, which is further connected to the discrimination circuit 7a, and the detectors 4a, 4b, 4c.
and 41a, 41b, 41c are all added to the adder 6b.
The calculation result is output to the discrimination circuit 7b. This makes it easy to adjust the mounting of the detector even if the knitted length is long.

In the two embodiments described above, the pitch of the detector is twice the pitch between the threads forming the stitches, but the present invention is not limited to this, and the pitch is set to an integral multiple of the pitch between the threads forming the stitches, or Of course, it is possible to implement the same method even if two or more detectors are provided at a pitch close to this.

Furthermore, in the above embodiments, in order to facilitate the explanation of the measurement principle, the width of the detector and the width of the gap between adjacent detectors are not explicitly stated, but the ratio of the former to the latter is preferably about twice. This is because it becomes difficult for harmonic components to be added to the detector signal. Here, if the pitch of the detector is twice the pitch of the warp yarns forming the stitches, then setting the ratio to about twice means that in the regular stitch section, as shown in Fig. 16, one Device element width WA (dotted chain line)
is approximately 4/3 times the pitch WB of the warp threads forming the stitch. In other words, it is the same as having about one and a half rows of warp stitches in WA.

Further, although the above two embodiments are transmission type in which the detection camera 1 and the light source 2 are provided with the knitted fabric X in between, the present invention is not limited to this. Of course, it is also possible to use a reflective type provided on the same side inside or outside of X.

Further, in the above description, the present invention is applied to flat knitting, but the present invention is not limited to this, and can also be applied to detecting flaws in various knitted fabrics such as milling knitting and double-sided knitting.

Effects of the Invention As is clear from the above explanation, the knitting flaw detection device according to the present invention can reliably detect the presence of needle knots. It is possible to stop the operation of a circular knitting machine, etc. before troubles due to fatal defects occur, contributing to the smooth running of the production process. Further, it is possible to detect hole-based flaws such as round holes, and different measures can be taken depending on the type of flaw, which is extremely useful for automating knitting machines.

[Brief explanation of drawings]

Fig. 1 is a perspective explanatory diagram showing an outline of the stitch flaw detection device according to the present invention, Fig. 2 is a block diagram thereof, and Figs. 3A to 3C are explanations showing the state in which regular stitches are detected. Figures 4A to 4E are explanatory diagrams showing the state in which a needle line is detected, Figure 5 is a graph showing the output of the subtraction section when a needle line is detected, and Figure 6 is a graph showing the output of the subtraction section when a needle line is detected.
The figure is an explanatory diagram showing the state in which a round hole is detected.
The figure is a graph showing the output of the adder when a round hole is detected, Fig. 8 is a block diagram showing another knitting flaw detection device, and Figs. 9A to 9E show the state in which needle marks are detected. Explanatory drawings: Fig. 10 is a graph showing the output of the addition/subtraction section when a needle thread is detected; Fig. 11 is an explanatory drawing showing the state in which a round hole is detected; Fig. 12 is a graph showing the output when a round hole is detected. A graph showing the output of the adder, Fig. 13 is an explanatory diagram of the angle formed between the optical axis of the light from the light source and the knitted fabric, Fig. 14 is a configuration diagram of a detector that can improve flaw detection sensitivity, and Fig. 15 The figure is a configuration diagram of a detection camera using the detector, and FIG. 16 is a diagram showing the stitches of the knitted fabric X and the ratio of the width of the detector to the gap width between adjacent detectors. 4a, 4b, 4c...detector, 6a...addition unit, 6
b... Subtraction section, 61a... Addition/subtraction section, W1 to W3, W
9~W18...Wale.

Claims (1)

[Scope of Claims] 1. A device for optically detecting flaws that do not form regular stitches in a knitted fabric, wherein the yarns are arranged at a pitch that is an integral multiple of the spacing between yarns that form the stitches, or at a pitch close to this. More than a book of photodetectors,
An arithmetic unit that adds and subtracts the output signals from each photodetector and a certain detection judgment value based on the presence or absence of a flaw are set, and a comparison of the magnitude of this detection judgment value and the output signal from the arithmetic unit detects a flaw. A discriminator for determining the presence or absence of a part.