CH703262A1 - METHOD AND DEVICE FOR THE OPTOELECTRONIC DETERMINATION OF CHARACTERISTICS OF A YARN. - Google Patents

Abstract

In the method according to the invention for the optoelectronic determination of features of a yarn (9), the yarn (9) is moved along its longitudinal direction by an optoelectronic yarn sensor device. The optoelectronic yarn sensor device takes successively a plurality of images (5.1, 5.2, ..., 5.n) of the yarn (9). From each individual image (5.1, 5.2, ..., 5.n), at least one feature of the yarn (9) is evaluated by means of image processing. The corresponding features evaluated from the individual images (5.1, 5.2, ..., 5.n) are linked together to form a result (7). Thus, characteristics of the yarn (9) such as e.g. its hairiness length distribution can be determined reliably and reproducibly.

Description

AREA OF EXPERTISE

The present invention is in the field of quality control in the textile laboratory. It relates to a method and a device for the optoelectronic determination of features, in particular the hairiness, of a yarn, according to the preambles of the independent claims. Such methods and devices are typically used in a textile laboratory.

STATE OF THE ART

In staple yarns that are spun from many fibers, individual fibers are not fully integrated into the actual yarn package. They partially protrude from the yarn body, which is known as the hairiness of the yarn. Hairiness is an important yarn property, which is why the textile industry has a great interest in its reliable and reproducible measurement and classification.

Many known methods and devices for determining the hairiness of yarns use a one-dimensional sensor line which is arranged transversely to the longitudinal direction of the yarn. An example of such a device is the USTER * ZWEIGLE HAIRINESS TESTER 5 device from the applicant, e.g. in “USTER® ZWEIGLE HAIRINESS TESTER 5 Application Handbook”, V1.0, September 2009. The yarn is illuminated with a laser beam directed towards the sensor line. The fibers protruding from the yarn partially shade the light so that the relevant sensor elements of the sensor line detect less light. The projections of the fiber shadows on the sensor line are classified into nine length intervals, with the surface of the yarn package defining the zero point. Further arrangements with linear sensor lines are described in US Pat. No. 4,948,260 A, US Pat. No. 5,875,419 A and EP-1 621 872 A2, whereby imaging or non-imaging methods can be used.

According to DE-19 818 069 A1, a digital camera with a two-dimensional sensor chip is used to determine the hairiness of yarns. An image of a piece of yarn is created with the camera, digitized and then analyzed. The image information is divided into information that represents the thread package itself and information that represents the fibers protruding from the thread package. This ensures that the thickness and thickness fluctuations of the yarn have no influence on the determination of the hairiness. The yarn is illuminated with evenly bright, diffuse backlight. The camera is connected to an evaluation device via a data line. The evaluation device contains a computer for processing the image data, a control device for controlling the yarn withdrawal and an output unit for the evaluated data. Other publications also suggest the use of a camera to determine the hairiness of yarns, e.g. DE-19 918 780 A1, US-5 654 554 A and EP-0 754 943 A2, the pattern of an optical Fourier transformation being recorded according to the latter.

DISCLOSURE OF THE INVENTION

It is an object of the present invention to provide a method and a device for optoelectronic determination of features of a yarn, which further improve the known methods and devices.

These and other objects are achieved by the method according to the invention and the device according to the invention as defined in the independent patent claims. Advantageous embodiments are specified in the dependent claims.

In the method according to the invention for optoelectronic determination of features of a yarn, the yarn is moved along its longitudinal direction by an optoelectronic yarn sensor device. The optoelectronic yarn sensor device successively records a plurality of images of the yarn. At least one characteristic of the yarn is evaluated from each individual image by means of image processing. The corresponding characteristics evaluated from the individual images are linked to form a result.

The link can include a statistical method. The number of recorded and evaluated images is, for example, between 10 and 10,000 and preferably between 100 and 1,000.

In a preferred embodiment, its hairiness length distribution is determined as a feature of the yarn. As a result, a histogram can then be output which is spanned by an abscissa along which the hair lengths are plotted in several classes, and an ordinate along which the number of the corresponding hair lengths in each class is plotted, and in which the number of corresponding hair lengths for each class is shown as columns with corresponding heights.

The image recording and the image evaluation can take place serially, d. H. the recording of the next image only begins after the evaluation of the previous image has been completed. Alternatively, the image acquisition and the image evaluation can take place in parallel, i. H. During the evaluation of the previous image, the recording of the next image begins.

An intermediate result can continuously be displayed, which is created by linking the corresponding features from all the images evaluated so far. This display can assist an operator in assessing whether enough images have been recorded and the measuring process can be ended.

The optoelectronic yarn sensor device preferably has an optoelectronic two-dimensional image sensor with a plurality of image elements arranged in a matrix and which are spaced b from one another in the longitudinal direction of the yarn. An exposure time t is used for taking a picture according to the formula

chosen, where p is a number with 1 p <10 and preferably. 1 ≤ p <3. The exposure time can be adjusted by periodically switching a light source on and off, i.e. H. stroboscopic lighting, and / or by periodically switching the image sensor on and off, d. H. an electronic camera shutter.

[0013] Even before the method is carried out, images can be periodically recorded and evaluated with regard to the presence or absence of a yarn, and if a yarn is present, the method is started automatically. During the method, images can additionally be evaluated for the presence of regular measurement conditions, and in the case of non-regular measurement conditions, the relevant image is not taken into account for the result and / or the method is terminated. One criterion for the existence of regular measurement conditions is e.g. B. a position of the yarn transverse to its longitudinal direction.

The inventive device for optoelectronic determination of features of a yarn includes an optoelectronic yarn sensor device with an optoelectronic image sensor having a plurality of picture elements, an evaluation device for evaluating images of the optoelectronic image sensor and a control device for controlling the device. The evaluation device and the control device are set up to carry out the method according to one of the preceding claims.

With the term "light" in this document is meant not only the visible range (VIS) of the electromagnetic spectrum, but also the adjacent ranges of ultraviolet (UV) and infrared (IR). Thanks to the inventive method and the inventive device characteristics of the yarn such as B. its hairiness length distribution can be determined reliably and reproducibly.

LIST OF DRAWINGS

[0016] The invention is explained in detail below with reference to the schematic drawings. Fig. 1 <sep> shows a device according to the invention in a perspective view. 2 shows schematically an optoelectronic yarn sensor device in cross section. 3 illustrates schematically the method according to the invention. FIG. 4 <sep> shows a flow diagram of an embodiment of the method according to the invention.

CARRYING OUT THE INVENTION

Fig. 1 shows from the outside a device 1 according to the invention for determining features of a yarn 9. One recognizes in this representation a substantially cuboid housing 11, various yarn guide elements 12-15 for guiding the yarn 9, a sensor cover 16 for an optoelectronic Yarn sensor device 3 and a roller cover 17 for a yarn conveyor device 4. The optoelectronic yarn sensor device 3 measures at least one parameter of the yarn 9, e.g. B. the length of fibers (“hairs”) that protrude from the yarn 9.

FIG. 2 shows a schematic representation of the optoelectronic yarn sensor device 3. The optoelectronic yarn sensor device 3 contains an illumination unit and an imaging unit. The lighting unit has at least one light source 31 and lighting optics 32. The light source 31 can e.g. a light emitting diode (LED) can be used. The light source 31 emits electromagnetic radiation in the visible, infrared and / or ultraviolet spectral range, which in this document is summarized under the term “light” for the sake of simplicity. The lighting optics 32 collimates the light emitted by the light source 31 onto a section of the yarn 9. The Köhler lighting is preferably used, which is well known from microscopy and which will not be discussed further here. The imaging unit has imaging optics 33 and an optoelectronic image sensor 34. The imaging optics 33 images the illuminated section of the yarn 9 onto the image sensor 34. The image sensor 34 is preferably a two-dimensional image sensor 34 with a plurality of picture elements (pixels) arranged in a matrix. Image sensors 34 of this type are commercially available in the form of integrated optoelectronic components in various technologies, for example very well known and widely used from digital cameras. Alternatively, a likewise known one-dimensional line sensor with a large number of pixels which are arranged on a straight line perpendicular to the longitudinal direction of the yarn can be used as image sensor 34. For the sake of simplicity, the illumination optics 32 and the imaging optics 33 are shown in FIG. 2 as simple lenses; Of course, however, more complex optical systems can be used which, in addition to lenses, contain other optical elements such as diaphragms or filters. The imaging is preferably carried out in transmission, i.e. H. without an inserted yarn 9, the light source 31 shines on the image sensor 34, and when yarn 9 is inserted, it shades at least part of the light. In this schematic illustration, the yarn 9 is drawn as a yarn body 91 with a circular cross section, from which individual fibers (“hairs”) 92 protrude essentially radially outward.

For the application of the hairiness length measurement, it is advantageous if a longitudinal axis 90 of the yarn 9 does not intersect an optical axis 30 of the optoelectronic yarn sensor device 3, but is spaced therefrom, for example by a distance of a = 2-6 mm and preferably by approx. a = 4 mm. This axial offset a is based on the assumption that (at least when a message is made about a certain yarn length) the yarn 9 is rotationally symmetrical and therefore the hairiness length values to be measured are the same along the yarn circumference. If this assumption of symmetry applies, only a half-space 93 defined by the yarn longitudinal axis 90 needs to be considered, and a half-space 94 complementary thereto can be disregarded. In the half-space 93 considered, the hairiness can be better, i. H. with a higher resolution.

The yarn 9 is moved along its longitudinal direction at a speed v through the optoelectronic yarn sensor device 3. The optoelectronic yarn sensor device 3 records a plurality n of images or “snapshots” of the yarn 9 one after the other. Such images 5.1, 5.2, 5.n are shown schematically in FIG. 3. In this representation the chronological sequence is indicated by a local one below the other; in reality, all images 5.1, 5.2, ..., 5.n are recorded in the same place, but at different times t. Each image 5.1, 5.2, ..., 5.n is evaluated individually in that at least one feature of the yarn 9 is determined by means of image processing methods known per se, e.g. B. a frequency distribution of the lengths of the hair 92. The frequency distributions in the individual images 5.1, 5.2, 5.n are shown in Fig. 3 by means of histograms 6.1, 6.2, 6.n. Each histogram 6.1, 6.2, ..., 6.n is spanned by an abscissa 61 along which the hair lengths in several, e.g. B. seven, classes are plotted, and an ordinate 62, along which the frequency (number) of the measured values in each class is plotted. The frequency values are shown for each class as columns with corresponding heights.

The evaluation of the images 5.1, 5.2, ..., 5.n is preferably carried out online, d. H. during the measurement process. In a serial variant of the method according to the invention, the recording of the next image 5.2 begins only after the evaluation of the previous image has been completed. In a parallel variant, however, the recording of the next image 5.2 is started as early as the evaluation of the previous image 5.1. Compared to the serial variant, the parallel variant has the advantage of a higher measuring speed and thus a shorter total measuring time, which, however, has to be bought at the expense of high computing power and a high data transfer rate.

The evaluation according to the invention of a plurality of images 5.1, 5.2, 5.n is used to obtain more precise and more reliable measurement results by means of statistical methods. For this purpose, the frequency values belonging to the same class from all images 5.1, 5.2, ..., 5.n can be added, averaged or linked with one another in some other way, from which a resulting histogram 7, spanned by corresponding axes 71, 72, arises . The resulting frequency values are subject to scatter, which are indicated with error bars in the resulting histogram 7 of FIG. 3 and should generally be smaller the more images 5.1, 5.2, ..., 5.n were taken into account in the evaluation, i.e. . H. the greater the number n. According to the present invention, the number n is e.g. B. between 10 and 10,000 and preferably between 100 and 1,000.

In the case of a one-dimensional image sensor 34 (see FIG. 2), in a first variant, several successive image lines can initially be combined to form a two-dimensional image 5.1, 5.2, 5.n, as shown in FIG. 3. The two-dimensional image 5.1, 5.2, 5.n is then expanded, as described above. In a second variant, each one-dimensional image is treated as described above for the two-dimensional images 5.1, 5.2, 5.n. An individual histogram 6.1, 6.2,6.n then either contains no bar or exactly one bar of height 1. Information about the frequency distribution of the hair lengths is only contained in the resulting histogram 7.

If the evaluation of the images 5.1, 5.2, ..., 5.n takes place online, the resulting histogram 7 can be displayed on a display device, e.g. B. a screen, and in addition, the frequency values and / or the scatter for the individual classes can be displayed numerically. An operator can use the displayed values to judge at his own discretion when the measuring process should be ended. Alternatively, with an online evaluation, as many images 5.1, 5.2, 5.n can be recorded and evaluated until the scatter in all classes has fallen below a specified threshold value. In a third alternative, a fixed, predetermined number n of images 5.1, 5.2, 5.n to be entered by an operator is recorded. In a fourth alternative, measurements are made during a fixed, predetermined total measuring time or to be entered by an operator. The total measuring time T results in each case from the number n recorded images 5.1, 5.2, 5.n and the image repetition rate f according to the following formula:

Thus, z. B. a total measurement of T = 60 s when n = 600 images are recorded with a refresh rate of f = 10 Hz.

When recording the yarn images 5.1, 5.2, ..., 5.n with a two-dimensional optoelectronic image sensor 34, the exposure time should be optimally selected. Two contradicting requirements must be weighed up: On the one hand, the exposure time should be as long as possible so that the images 5.1, 5.2, 5.n have a high signal-to-noise ratio, and on the other hand, it should be so short that the images 5.1, 5.2, ..., 5.n are not smeared as a result of the yarn movement, but are sharp. Applied to the specific case of a digital two-dimensional image sensor 34 with a large number of discrete picture elements ("pixels") arranged in a matrix form, the second requirement states that at a given yarn speed, an image point only has a very small number p of pixels (1 ≤ p <10 and preferably 1 ≤ p <3), ideally just p ≈1 pixel. If v is the yarn speed and b is the pixel spacing in the longitudinal direction of the yarn, this results in the maximum exposure time

In an example with v = 400 m / min, b = 10 (.im and p = 2 one obtains t ≤ 3 microseconds. In addition, in order to meet the first requirement, the intensity of the lighting and the sensitivity of the image sensor 34 can be selected so that a sufficiently good image signal results with such a short exposure time t. The exposure time t can be determined by periodically switching the light source 31 on and off, ie stroboscopic lighting, and / or by periodically switching the image sensor 34 on and off, ie an electronic camera shutter.

The evaluation of the images 5.1, 5.2, 5.n is preferably carried out in a (not shown in the drawings) central control device of the device 1. The control device contains a processor, preferably a digital signal processor (DSP). The processor is connected to the optoelectronic yarn sensor device 3 and to the conveyor device 4 (see FIG. 1). A program stored in the processor controls their switching on and off, the intensity of the lighting, the retrieval and evaluation of the recorded images 5.1, 5.2, 5.n, the yarn conveying speed v and / or other parameters. As a result, it controls the sequence of the determination of the features of the yarn 9 in the device 1. FIG. 4 shows an example of a flow chart which can be the basis of the program and which defines an embodiment of the method according to the invention.

The control device can be programmed such that the measuring process is started automatically as soon as a yarn 9 is inserted into the optoelectronic yarn sensor device 3. The optoelectronic yarn sensor device 3 can determine whether this is the case. In a standby mode, it periodically records 101 images, and the control device evaluates these with regard to the presence or absence of a yarn 9. If this evaluation shows 102 that a yarn 9 has been inserted into the optoelectronic yarn sensor device 3, the device 1 switches to the measuring mode. The measuring process is started 103, which includes switching on the conveyor device 4, a yarn traversing device, a yarn suction device, etc., among other things.

As soon as the yarn 9 runs stably through the optoelectronic yarn sensor device 3, the actual measurement and evaluation 104 can begin, as described on the occasion of FIG. 3. If several images 5.1, 5.2, 5.n are present 105, they are evaluated 106 to produce a result, for example the resulting histogram 7. If a criterion for ending the measurement process - e.g. B. shutdown by an operator, falling below a threshold value for the scatter, calibrating a predetermined number n of images 5.1, 5.2, 5.n or reaching a predetermined total measuring time T - is fulfilled 107, a result is output 108 and the measuring process is ended 109. The device 1 can return to the standby mode or be switched off completely.

During the measurement mode, the device 1 can check 110 whether there are regular measurement conditions as provided for the measurement. This would not be the case, for example, if the yarn 9 or its image were in a position other than that shown in FIGS. 2 and 3. The optoelectronic yarn sensor device 3 can determine such a yarn run error. With regular measurement conditions, the measurement process proceeds as described above. However, if the measurement conditions for a measurement are not regular, the image recorded is not taken into account. Depending on the programming of the control device, either a new image is recorded or the entire measurement process is aborted and a warning message is output 111. Whether the measurement process is aborted can depend on the extent of the deviation from the regular measurement conditions or on a number of previously determined irregular conditions Measurement conditions are made dependent.

Of course, the present invention is not limited to the embodiments discussed above. With knowledge of the invention, the person skilled in the art will be able to derive further variants which also belong to the subject matter of the present invention as defined in the independent claims.

REFERENCE LIST

1 <sep> device 11 <sep> housing 12-15 <sep> guide elements for the yarn 16 <sep> sensor cover 17 <sep> roller cover 2 <sep> control device 21 <sep> processor 22 <sep> printed circuit board 3 < sep> optoelectronic yarn sensor device 30 <sep> optical axis of yarn sensor device 31 <sep> light source 32 <sep> illumination optics 33 <sep> imaging optics 34 <sep> image sensor 4 <sep> conveyor device 5 <sep> image of a yarn section 6 <sep> histogram with Length distribution from a single image 61 <sep> abscissa 62 <sep> ordinate 7 <sep> resulting histogram 71 <sep> abscissa 72 <sep> ordinate 9 <sep> yarn 90 <sep> longitudinal axis 91 <sep> package 92 <sep> Yarn hairs 93, 94 <sep> half spaces a <sep> Axis offset between the optical axis and the longitudinal axis of the yarn

Claims (10)

1. Method (1) for the optoelectronic determination of features of a yarn (9), the yarn (9) being moved along its longitudinal direction by an optoelectronic yarn sensor device (3), and the optoelectronic yarn sensor device (3) images (5.1, 5.2, 5 .n) of the yarn (9), characterized in that the optoelectronic yarn sensor device (3) successively records (104) a plurality of images (5.1, 5.2, 5.n) of the yarn (9), from each one Image (5.1, 5.2, 5.n) at least one feature of the yarn (9) is evaluated (104) by means of image processing, and the corresponding features evaluated from the individual images (5.1, 5.2, 5.n) to produce a result (7 ) are linked together (106). 2. The method of claim 1, wherein the link (106) includes a statistical method. 2. The method of claim 1, wherein the link (106) includes a statistical method. 3. The method according to any one of the preceding claims, wherein its hairiness length distribution is determined as a feature of the yarn (9). 3. The method according to any one of the preceding claims, wherein the number of recorded and evaluated images (5.1, 5.2, 5.n) is between 10 and 10,000 and preferably between 100 and 1,000. 4. The method according to claim 3, wherein a histogram (7) is output as a result, which is represented by an abscissa (71) along which the hair lengths are plotted in several classes, and an ordinate (72) along which the number of corresponding hair lengths is applied in each class, is stretched, and in which the number of corresponding hair lengths for each class is shown as columns with corresponding heights. 4. The method according to any one of the preceding claims, wherein as a feature of the yarn (9) its hairiness length distribution is determined. 5. The method according to any one of the preceding claims, wherein the image recording (104) and the image evaluation (104) take place serially, i.e. the recording of the next image (5.2) only begins after the evaluation of the previous image (5.1) has been completed. 5. The method according to claim 4, wherein a histogram (7) is output as a result, which is represented by an abscissa (71) along which the hair lengths are plotted in several classes, and an ordinate (72) along which the number of the corresponding hair lengths is applied in each class, is stretched, and in which the number of corresponding hair lengths for each class is shown as columns with corresponding heights. 6. The method according to any one of claims 1-4, wherein the image recording and the image evaluation take place in parallel, d. H. Even during the evaluation of the previous image (5.1), the recording of the next image (5.2) is started. 6. The method according to any one of the preceding claims, wherein the image recording (104 and the image evaluation (104) take place serially, i.e. the recording of the next image (5.2) only begins after the evaluation of the preceding image has been completed: (5.1). 7. The method according to any one of the preceding claims, wherein before the execution of the method, images are periodically recorded (101) and evaluated (101) with regard to the presence or absence of a yarn (9), and if a yarn (9) is present, the method is automatic is started (103). 7. The method according to any one of claims 1-5, wherein the image recording and the image evaluation take place in parallel, i.e. Even during the evaluation of the previous image (5.1), the recording of the next image (5.2) is started. 8. The method according to one of the preceding claims, wherein images (5.1, 5.2, ..., 5.n) are additionally evaluated for the presence of regular measurement conditions (104) and, in the case of non-regular measurement conditions, the relevant image is not taken into account for the result and / or the method is terminated (109). 8. The method according to any one of the preceding claims, wherein an intermediate result is continuously displayed, which is created by linking (106) the corresponding features from all previously evaluated images. 9. The method according to spoke 8, wherein a criterion for the existence of regular measurement conditions (110) is a position of the yarn (9) transversely to its longitudinal direction. 9. The method according to any one of the preceding claims, wherein the optoelectronic yarn sensor device (3) has an optoelectronic two-dimensional image sensor (34) with a plurality of image elements arranged in the form of a matrix which are spaced b from one another in the longitudinal direction of the yarn and for recording (104) an image (5.1, 5.2, 5.n) an exposure time t according to the formula is chosen where p is a number with 1 p p <10 and preferably. 1 ≤ p <3. 10. The method according to claim 9, wherein the exposure time t by periodically switching on and off a light source (31), i. stroboscopic lighting, and / or by periodically switching the image sensor (34) on and off, d. H. an electronic camera shutter. 11. The method according to any one of the preceding claims, wherein images are periodically recorded (101) and evaluated (101) with regard to the presence or absence of a yarn (9), and if a yarn (9) is present, the method is automatic is started (103). 12. The method according to any one of the preceding claims, wherein images (5.1, 5.2, ..., 5.n) are additionally evaluated for the presence of regular measurement conditions (104) and, in the case of non-regular measurement conditions, the relevant image is not taken into account for the result and / or the method is terminated (109). 13. The method according to claim 12, wherein a criterion for the existence of regular measurement conditions (110) is a position of the yarn (9) transversely to its longitudinal direction. 14. Device (1) for optoelectronic determination of features of a yarn (9), comprising an optoelectronic yarn sensor device (3) with an optoelectronic image sensor (34) which has a large number of picture elements, an evaluation device for evaluating images (5.1, 5.2 , 5.n) of the optoelectronic image sensor and a control device (2) for controlling the device (1), characterized in that the evaluation device and the control device are set up to carry out the method according to one of the preceding claims. 1. Method (1) for the optoelectronic determination of features of a yarn (9), the yarn (9) being moved along its longitudinal direction by an optoelectronic yarn sensor device (3), and the optoelectronic yarn sensor device (3) images (5.1, 5.2, 5 .n) of the yarn (9), characterized in that the optoelectronic yarn sensor device (3) successively records (104) a plurality of images (5.1, 5.2, 5.n) of the yarn (9), from each one Image (5.1, 5.2, 5.n) at least one feature of the yarn (9) is evaluated (104) by means of image processing, and the corresponding features evaluated from the individual images (5.1, 5.2, 5.n) to produce a result (7 ) are linked together (106). 10. Device (1) for optoelectronic determination of features of a yarn (9), comprising an optoelectronic yarn sensor device (3) with an optoelectronic image sensor (34) which has a large number of picture elements, an evaluation device for evaluating images (5.1, 5.2, 5.n) of the optoelectronic image sensor and a control device (2) for controlling the device (1), characterized in that the evaluation device and the control device are set up to carry out the method according to one of the preceding claims.