Knitting machine having a monitoring device and method for monitoring needles during operation of the knitting machine

EP4709916A1Pending Publication Date: 2026-03-18PROTECHNA HERBST
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
EP · EP
Patent Type
Applications
Current Assignee / Owner
Filing Date
2024-04-18
Publication Date
2026-03-18

AI Technical Summary

Technical Problem

Existing methods for monitoring needles in knitting machines require reference data for error detection, which can be cumbersome and inefficient, and do not allow for direct checking of needle positions without comparison.

Method used

A monitoring device with a camera and light source, capable of capturing high-frame-rate images of needles without reference information, allowing for direct identification of needle positions and faults by measuring brightness values and determining median distances between needles.

Benefits of technology

Enables quick and reliable detection of faulty or missing needles without the need for reference data, improving operational efficiency and allowing for automated adjustment and independent operation across different knitting machine modes.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure EP2024060645_21112024_PF_FP_ABST
    Figure EP2024060645_21112024_PF_FP_ABST
Patent Text Reader

Abstract

In order to monitor needles (3) during the operation of a knitting machine, the knitting machine is provided with a monitoring device having a camera (4), a light source (5) and a control unit (11) having a camera control. The light source (5) and the camera (4) are arranged relative to one another such that light rays (6) incident on at least one needle (3) are emitted from the light source (5), and light rays (7) reflected from the needles (3) are detected by the camera (4). The camera (4) has a surface camera element. Images of an image area (9), which corresponds to at least one needle (3) or at least one part of the needle (3), are acquired, wherein the image data are measured brightness values of the reflected light rays (7). The frame rate fK, i.e. the number of images per unit time, and the intensity of the reflected light rays (7) of the light source (5) are selected to be high enough that, in at least two consecutive images, all of the needles (3) of the image area (9), which comprises at least five needles (3), are imaged from the image data with a minimum separation. From the image data, the position of each needle (3) is identified on the image, and from the positions of the needles (3), a defective needle (3) is identified.
Need to check novelty before this filing date? Find Prior Art

Description

[0001] DESCRIPTION

[0002] Knitting machine with a monitoring device and method for monitoring needles during operation of the knitting machine

[0003] The invention relates to a knitting machine with a monitoring device and a method for monitoring needles during operation of the knitting machine according to the preambles of the independent claims.

[0004] A method for optical quality control of needles on knitting machines is known from DE 40 14 661 A1. To detect gross defects, an image of the needle zone to be monitored is projected onto a semiconductor matrix server synchronously with the needle frequency. The analog brightness signals of the discrete pixels are digitized and compared with reference data.

[0005] In the method for inspecting needles using a light scanning system described in DE 199 24924 A1, light reflected from the needles from a first optical fiber passes into a second optical fiber. Intact, correctly positioned needles generate a basic signal sequence on a receiver. During operation, a signal sequence is recorded and compared with the basic signal sequence.

[0006] US 2022 / 0349097 A1 discloses using a camera module to monitor a knitting machine based on the surface of the knitted fabric, which compares images of the same vertical line. DE 102 23 754 A1 describes detecting defects in a knitted fabric by comparing, for example, information captured by a camera from an illuminated fabric web with stored information.

[0007] A camera with area camera elements for monitoring a moving web of material is known from DE 10 2015 119444 B4.

[0008] Many methods for monitoring the needles of a knitting machine involve recording reference data in advance, which is then used during operation to detect errors.

[0009] The object of the invention is to develop a knitting machine with a monitoring device and a method for monitoring needles during operation of the knitting machine, which enables the direct checking of the needles without comparison with reference information.

[0010] The problem is solved by the independent claims.

[0011] A knitting machine according to the invention comprises a monitoring device for monitoring needles during operation of the knitting machine. The monitoring device is equipped with a camera, a light source, and a control unit with a camera controller. The light source and the camera are arranged relative to one another such that light rays striking at least one needle are emitted from the light source, and light rays reflected by the needles are detected by the camera.

[0012] The camera has an area-scan camera element. The camera controller is configured to capture images of an image area corresponding to at least one needle or at least part of the needle. The image data are measured brightness values ​​of the reflected light rays. The camera has such a high frame rate, and the reflected light rays from the light source have such a high intensity, that all needles in the image area are imaged from the image data at a minimum distance from each other in at least two consecutive images.

[0013] The image area comprises at least five needles. The control unit is configured to identify the position of each needle in the image from the image data and to identify a faulty needle from the positions of the needles.

[0014] The monitoring device makes it possible to generate at least two consecutive images in which all needles in the image area are clearly depicted from the image data. Generating such images makes it possible to identify and verify the positions and thus the distances between all visible needles. No reference information is required for this.

[0015] In one embodiment, the camera is designed to be operated with a frame rate fK that is greater than a maximum possible needle rate fN, ie the number of needles per time at which the needles pass the camera.

[0016] By adjusting the camera frame rate to the highest possible needle rate of the knitting machine, the monitoring device can be used independently of the knitting machine's operating mode.

[0017] In one embodiment, the camera controller is configured to combine pixels of the area-scan camera element for reading the image data of the image area. This increases the readout speed.

[0018] In one embodiment, the camera control is designed to select a selection field of selected pixels for the image area for reading out the image data.

[0019] This also increases the reading speed.

[0020] In one embodiment, the camera control is designed to move the selection field for fine adjustment of the image area.

[0021] Moving the selection field electronically is a much simpler adjustment than using a mechanical mechanism, such as adjusting screws. Moving the selection field using the control unit allows for automated adjustment.

[0022] In one embodiment, the camera and the light source are arranged such that the image area captures at least five expelled needles in the region of the reversal point.

[0023] This allows for a lower height of the image area and thus faster reading and evaluation of the image data.

[0024] In one embodiment, the distance of the camera to the needles is 5 cm to 40 cm and / or the distance of the light source to the needles is 5 cm to 40 cm.

[0025] For the operation of the knitting cylinder and the camera mount, it is advantageous to position the camera at a distance of at least 5 cm or more from the needles. In one embodiment, the light source is designed to emit the light rays striking the needles as a convergent beam.

[0026] Due to the convergent beam of rays, the intensity of the light rays striking the surface element of the camera is increased in the case of a good needle and decreased in the case of a broken needle due to the rough fracture surface.

[0027] In one embodiment, the control unit is designed to determine a median value for the distances between the imaged needles from the image data of a plurality of images and a moving average as a standard distance from the median values ​​of the images, and to compare each individual determined distance between two needles with the standard distance in order to determine a faulty or missing needle.

[0028] The formation of median values ​​and an updated standard distance enables a quick and reliable detection of a faulty or missing needle.

[0029] The methods according to the invention have corresponding features and advantages.

[0030] Further features and benefits are set out in the following description.

[0031] The invention is further explained with reference to examples schematically illustrated in the drawing. Figure 1 shows a schematic view of a circular knitting machine with a monitoring device;

[0032] Figure 2 is a simplified view of a row of needles with an image area identified;

[0033] Figure 3 is a block diagram of the monitoring device;

[0034] Figure 4 a, b, c show consecutive images with image data of an image area with five needles in the area of ​​ascent or descent of the needles;

[0035] Figure 5 is an image of an image area with seven needles in the area of ​​the reversal point.

[0036] Figure 1 shows a circular knitting machine 1 with a knitting cylinder 2 and stitch-forming needles 3. The thread feed is left out in Figure 1 for reasons of clarity.

[0037] The circular knitting machine 1 has a monitoring device provided with a camera 4 and a light source 5. The light source 5 and the camera 4 are arranged together in such a way that light beams 6 striking the needles 3 are emitted by the light source 5 and light beams 7 reflected by the needles 3 are detected by the camera 4. The angle between the striking and reflected light beams 6, 7 is approximately 60°. The camera 4 is arranged approximately at the height of the needles 3 and the light source 5 above them. The camera 4 and the light source 5 are attached to a common holder 8 on the circular knitting machine 1. The distance of the camera 4 to the needles 3 and / or the distance of the

[0038] Light source to needles 3 is 5 cm to 40 cm.

[0039] Figure 2 shows a simplified row of needles 3, with an image area 9 marked with five needles 3. Also marked is an area 10 within the image area 9 below the needle heads 3a, which may be masked out, e.g., with a digital mask.

[0040] As shown in the block diagram of Figure 3, the monitoring device comprises a control unit 11 connected to the camera 4 and having a camera controller. The camera controller is integrated into the control unit 11.

[0041] In an alternative, the camera control is located in camera 4.

[0042] The camera control system includes, for example, a frame grabber card. The control unit 11 is equipped with a microprocessor. The camera control system is designed to mask out any pixels that are located off the path traversed by the needle heads 3a using a digital mask. In this example, the pixels in area 10 are masked out.

[0043] In an alternative, the control unit 11 with the camera control is designed as a hardware unit.

[0044] A power supply 12 is connected to the control unit 11 and the light source 5.

[0045] The control unit 11 includes a switch 13 for generating a stop signal, which is connected to a machine control 14. When the switch 13 is switched by the control unit 11, a stop signal is sent to the machine control 14 and the circular knitting machine 1 is stopped.

[0046] In an alternative, the control unit 11 is connected to the machine control via a bus system, e.g. a CAN bus or an Ethernet bus, via which a stop signal is sent.

[0047] A display device 15 is connected to the control unit 11, for example, for displaying images selected by the control unit 11. Since the images are not perceivable at the frame rate fK of the camera 4, only selected images are displayed, for example, images selected every 1 / 30 of a second. The display device 15 also shows a distance determined by the control unit 11, for example, a number of needles 3, between a specific needle 3 and a specific position in the knitting cylinder 2. The specific needle 3 is, for example, a needle 3 identified as defective or missing. The specific position is, for example, a repair position.

[0048] The camera 4 has an area camera element. The camera control is designed to capture an image of an image area 9 with at least five extruded needles 3.

[0049] In alternatives, the image area 9 includes 3 to 20 needles.

[0050] The image data of an image are the brightness values ​​of the rays 7 reflected by the needles 3, which are measured per image point, ie per pixel, of the area camera element.

[0051] Figures 4a, 4b, 4c show three consecutive images of the image area 9 schematically shown in Figure 2. The camera 4 has such a high frame rate fK and the reflected light rays 7 of the light source 5 such a high intensity that in the three consecutive images of Figures 4a, 4b, 4c, all needles 3 of the image area 9 are imaged from the image data with a minimum distance.

[0052] In this example, the camera control is designed to capture images of an image area with images of at least five needle heads 3a.

[0053] In the images of Figures 4a, 4b, and 4c, the five needles 3 are recognizable as images of the needle heads 3a, i.e., as bright spots corresponding to the needle heads 3a. The images of the needle heads 3a increase from bottom left to top right.

[0054] A minimum distance between the images, e.g., between the right flank of a bright spot and the right flank of the following bright spot, is present when the distance between the images is at least three times greater than the width of the images. This minimum distance ensures sufficient distance between the images to detect a missing or broken needle 3.

[0055] Image area 9, for example, comprises 256 pixels in a row and 200 pixels in a column. The spacing between the images is approximately 50 pixels. The area of ​​each image is approximately 10 pixels, and the distance between the edges of the images is approximately 40 pixels. The spacing between the images is approximately five times greater than their width. The spacing between the images corresponds to the spacing between the three pins.

[0056] The minimum spacing is, for example, 20 to 50 pixels, with an image width of 5 to 15 pixels. Based on the different images of the needle heads 3a of the five needles 3, it can be seen that over the course of the three images, Figures 4a, b, and c, the images move toward the bottom left.

[0057] The light source 5 has an optical system, which generates a light spot that essentially illuminates only the needles 3 of the image area 9. In this example, the light spot essentially illuminates the needle heads 3a of the needles 3. The light spot is, for example, round and has a diameter of, for example, 5 mm.

[0058] In alternatives, the light source 5 has an optic that produces an elliptical or a rectangular light spot.

[0059] The light source 5 is a spotlight, in particular an LED spotlight with a power of at least 3 watts.

[0060] In an alternative, the light source 5 is provided as an LED spotlight with a power of 12 watts.

[0061] In an alternative, the light source 5 is designed as a stroboscopic light source.

[0062] The light source 5 is configured with its optics to emit the light rays 6 striking the needles 3 as a convergent beam. The convergent beam striking the needles 3 is reflected more strongly by the convex needle heads 3a than a parallel beam. Furthermore, light rays 6 of a convergent beam are scattered more strongly by a rough fracture surface of a broken needle 3. Alternatively, the camera 4 is equipped with a telecentric lens.

[0063] The frame rate fK is the number of images per time that camera 4 captures.

[0064] The camera control is designed to operate the camera 4 with a frame rate fK that is greater than a needle rate fN of the circular knitting machine 1, ie the number of needles 3 per time at which the needles 3 pass the camera 4.

[0065] At a speed of the circular knitting machine 1 of 1000 needles 3 per second, ie a needle rate fN of 1 kHz, a slightly larger frame rate fK e.g. of 1.1 kHz or e.g. of 1.5 or e.g. of 2 kHz is used.

[0066] The camera control is configured to set the exposure time of camera 4, i.e., the time during which the surface element of camera 4 measures the intensity of the incident light as brightness values, to 2 to 15% of the frame interval. In this example, the exposure time is set to 3 to 5% of the frame interval. At a frame rate of 2 kHz, the frame interval is 0.5 ms.

[0067] The camera control is preferably designed to operate the camera 4 with a frame rate fa that is greater than a maximum possible needle rate fN of the circular knitting machine 1.

[0068] In an alternative, the speed of circular knitting machine 1 is 6000 needles per second, for example, and the needle rate fN is 6 kHz. A frame rate of 7 kHz, for example, is used.

[0069] The area-scan camera element of camera 4 is equipped with approximately 2500 x 2000 pixels. To read out the image data, i.e., the brightness values, of image area 9, the camera controller is configured to binnage pixels from the area-scan camera element. This process is called binning. The pixels are binned in sets of 4 by 4, reducing the number of pixels to be read out to 600 x 500.

[0070] The camera control is designed to select a selection field of selected pixels for the image area 9. A selection field is referred to as a region of interest (roi).

[0071] The selection field, for example, has a size of 256 x 30 to 200 pixels. The camera control is designed to shift the image area 9 for fine adjustment, i.e., for precise alignment of the image area 9 to the needles 3. This allows, after mounting the camera 4 and the light source 5, the image area 9 to be precisely aligned to the five desired needles 3 of the knitting cylinder 2.

[0072] In an alternative, the area camera element of camera 4 is provided with 1300 x 1100 pixels, for example.

[0073] In another alternative, the area camera element has 256 x 200 pixels.

[0074] The control unit 11 is designed to identify the position of each needle 3 from the image data, ie the brightness values ​​of the pixels, and to identify a faulty needle 3 from the positions of the needles 3.

[0075] The control unit 11 is designed to determine a median value from the image data of several images for the distances between the imaged needles 3 and a moving average from the median values ​​of the images as a standard distance and to compare each individual determined distance between two needles 3 with the standard distance in order to determine a faulty or missing needle 3.

[0076] During operation of the circular knitting machine 1, the light source 5 and the camera 4 are arranged relative to one another for monitoring needles 3 by a monitoring device such that light beams 6 are emitted by the light source 5 onto at least five needles 3, and light beams 7 reflected by the needles 3 are captured by the camera 4 with a surface element. The camera control system captures images of an image area 9, wherein the image area 9 corresponds to at least five needles 3. The image data are measured brightness values ​​of the light beams 7 reflected by the five needles 3.

[0077] In this example, the image data are the brightness values ​​of the light rays 7 reflected by the five needle heads 3a.

[0078] The images are recorded by the camera 4 at such a high frame rate f« and are recorded with such a high intensity of the light rays 6 emitted by the light source 5 and the light rays 7 reflected by the needle heads 3a of the needles 3 that in at least two consecutive images all needles 3 are imaged by their needle heads 3a of the image area 9 from the image data with a minimum distance.

[0079] The light source 5 emits the light rays 6 striking the needles 3, specifically the needle heads 3a, as a convergent beam. The camera control operates the camera 4 at a frame rate fK that is greater than the maximum possible needle rate fN of the circular knitting machine 1 at which the needles 3 move past the camera 4. The frame rate fK is, for example, 2 kHz at a maximum needle rate fN of 1.5 kHz and 7 kHz at a maximum needle rate fN of 6 kHz.

[0080] The camera control combines pixels of the area scan camera element for reading the image data of image area 9 and / or selects a selection field of selected pixels for image area 9. The camera control may move the selection field for fine adjustment of image area 9.

[0081] The control unit 11 identifies the position of each needle 3 from the image data and identifies faulty needles from the positions of the needles 3.

[0082] The control unit 11 determines a median value from the image data of several images for the distances between the imaged needles 3 and a moving average from the median values ​​of the images as the standard distance. In order to determine a faulty or missing needle 3, each individual determined distance between two needles 3 is compared with the standard distance.

[0083] The position of a needle 3 in an image is determined by the control unit 11 as follows: The analog brightness values ​​of the pixels are converted by the camera 4 into values ​​0 to 255, i.e. digitized. A minimum brightness value is determined which indicates a needle 3 or a pin head 3a, for example the value 50. For evaluation, either a value 0 is determined column by column if the highest value measured in the column is below 50, or a value 1 is determined if the highest value measured in the column is 50 to 255. The position of the first needle 3 in the image is determined at the first determined value of 1, e.g. moving from left to right in the image. If the image extends over 10 pixels, for example, the next 9 columns also have the value 1. Then, between the images, the values ​​of the columns are 0 up to the next column with the value 1, at which column the position of the second needle 3 is determined.

[0084] The distance between the first determined values ​​of the columns of 1 is evaluated by the control unit 11 as the distance between the needles 3. The control unit 11 uses the image data of several images to determine a median value for the determined distances between the depicted needles 3 of an image, and then uses the median values ​​of the images to determine a moving average as the standard distance. For example, the last 50 images are considered. Alternatively, the last 20 to 100 images are considered.

[0085] To detect a faulty or missing needle 3, the control unit 11 compares each individual determined distance between two needles 3 with the standard distance.

[0086] A faulty or missing needle 3 is detected when the distance exceeds a threshold.

[0087] A faulty needle 3 is, for example, a bent needle 3. A missing needle 3 is, for example, a needle in which the needle head 3a is broken off.

[0088] A first threshold for the distance is 1.2 times the standard distance. If the first threshold is exceeded, a faulty needle 3 is detected. A second threshold for the distance is 1.5 times the standard distance. If the second threshold is exceeded, a missing needle 3 is detected.

[0089] The position of a needle 3 with respect to the knitting cylinder 2 is determined, for example, during installation by the control unit 11 in relation to a specific position in the knitting cylinder 2, e.g. to a specific lock S of the knitting cylinder

[0090] 2 or to a repair position as the number of needles 3. When the circular knitting machine 1 is operating, the needles 3 shown in the illustrations are removed from the needles by the maximum number of needles with each revolution.

[0091] 3 of the knitting cylinder 2 is counted down. The current distance, e.g., to a repair point if a faulty needle 3 has been detected, is also shown on the display device 15.

[0092] In an alternative, the position of a needle 3 relative to the knitting cylinder 2 is determined by the control unit 11 by counting the needles 3 relative to a position sensor 16 of the knitting cylinder 2.

[0093] Figure 5 shows an image of a second example, which corresponds to the first except for the following features. The camera 4 and the light source 5 are arranged such that the image area 9 encompasses seven extruded needles 3 in the region of the reversal point.

[0094] A fracture needle can be seen at the fifth point coming from the left.

[0095] In this example, the minimum brightness value indicated by a pin 3 or a pin head 3a is determined, for example, as 220. The brightest pixel of the breakpoint pin lies below the minimum brightness value. The breakpoint pin is ignored when determining the positions and distances by control unit 11. The distance between the adjacent pins 3 is a factor of 1.5 greater than the standard distance determined from the median values. Control unit 11 detects a faulty pin 3 and sends a stop signal to machine control 14 via switch 13.

[0096] List of reference symbols . Circular knitting machine . Knitting cylinder . Needle a Needle head . Camera . Light source . Light beam . Light beam . Holder . Image area 0. Area ^ Control unit 2. Power supply 3. Switch 4. Machine control 5. Display device 6. Position sensor Lock

Claims

CLAIMS 1. Knitting machine with a monitoring device for monitoring needles (3) during operation of the knitting machine, wherein the monitoring device is provided: with a camera (4), with a light source (5) and with a control unit (11) with a camera controller, wherein the light source (5) and the camera (4) are arranged relative to one another in such a way that light rays (6) striking at least one needle (3) are emitted by the light source (5) and light rays (7) reflected by the needles (3) are captured by the camera (4), wherein the camera (4) has an area camera element, and wherein the camera controller is designed to capture images of an image region (9) corresponding to at least one needle (3) or at least a part of the needle (3), wherein the image data are measured brightness values ​​of the reflected light rays (7), characterized in that the camera (4) has such a high frame rate f«, ieNumber of images per time, and the reflected light rays (7) of the light source (5) have such a high intensity that in at least two consecutive images all needles (3) of the image area (9) from the image data are imaged with a minimum distance, wherein the image area (9) comprises at least five needles (3), and the control unit (11) is designed to identify the position of each needle (3) on the image from the image data and to identify a faulty needle (3) from the positions of the needles (3).

2. Knitting machine with a monitoring device according to claim 1, characterized in that the camera control is designed to operate the camera (4) with a frame rate fK which is greater than a maximum possible needle rate fbj, ie number of needles (3) per time at which the needles (3) pass the camera (4).

3. Knitting machine with a monitoring device according to claim 1 or 2, characterized in that the camera control is designed to combine pixels of the area camera element for reading out the image data of the image area (9).

4. Knitting machine according to one of claims 1 to 3, characterized in that the camera control is designed to select a selection field of selected pixels for the image area (9) for reading out the image data.

5. Knitting machine according to claim 4, characterized in that the camera control is designed to move the selection field for fine adjustment of the image area (9).

6. Knitting machine with a monitoring device according to one of claims 1 to 5, characterized in that the camera (4) and the light source (5) are arranged so that the image area (9) detects at least five driven-out needles (3) in the area of ​​the reversal point.

7. Knitting machine with a monitoring device according to one of claims 1 to 6, characterized in that the distance of the camera (4) to the needles (3) is 5 cm to 40 cm and / or the distance of the light source (5) to the needles (3) is 5 cm to 40 cm.

8. Knitting machine with a monitoring device according to one of claims 1 to 7, characterized in that the light source (5) is designed to emit the light rays (6) striking the needles (3) as a convergent beam.

9. Knitting machine with a monitoring device according to one of claims 1 to 8, characterized in that the control unit (11) is designed to determine a median value from the image data of a plurality of images for the distances between the imaged needles (3) and a moving average value as a standard distance from the median values ​​of the images and to compare each individual determined distance between two needles (3) with the standard distance in order to determine a faulty or missing needle (3).

10. Method for monitoring needles (3) during operation of a knitting machine with a monitoring device, wherein the monitoring device is provided with a camera (4), with a light source (5) and with a control unit (11) with a camera control, wherein the light source (5) and the camera (4) are arranged relative to one another in such a way that light rays (6) striking at least one needle (3) are emitted by the light source (5) and light rays (7) reflected by the needles (3) are detected by the camera (4), wherein the camera (4) has an area camera element, and wherein images of an image area (9) are captured by the camera control, wherein the image area (9) corresponds to at least one needle (3) or at least a part of a needle (3), wherein the image data are measured brightness values ​​of the light rays (7) reflected by the at least one needle (3), characterized in that images are captured by the camera (4) with such a high frame rate f« and with such a high intensity by the light rays (6) emitted by the light source (5) and light rays (7) reflected by the needles (3) that in at least two consecutive images, all needles (3) of the image area (9) are imaged from the image data with a minimum distance, wherein the image area (9) comprises at least five needles (3), and the control unit (11) determines the position of each needle (3) from the image data and the positions of the needles (3),faulty needles (3) are identified., 11. Method for monitoring needles (3) during operation of a knitting machine according to claim 10, characterized in that the camera control operates the camera (4) at a frame rate f« which is greater than a maximum possible rate fj at which the needles (3) pass the camera (4).

12. Method for monitoring needles (3) during operation of a knitting machine according to claim 10 or 11, characterized in that pixels of the area camera element are combined by the camera control for reading out the image data of the image area (9).

13. Method according to one of claims 10 to 12, characterized in that a selection field of selected pixels for the image area (9) is selected by the camera control for reading out the image data.

14. Method according to claim 13, characterized in that the selection field is moved by the camera control for fine adjustment of the image area (9).

15. Method for monitoring needles (3) during operation of a knitting machine according to one of claims 10 to 14, characterized in that the camera (4) and the light source (5) are arranged such that at least five driven-out needles (3) are detected in the vicinity of the reversal point in the image area (9).

16. Method for monitoring needles (3) during operation of a knitting machine according to one of claims 10 to 15, characterized in that the distance of the camera (4) to the needle section is 5 cm to 40 cm and / or the distance of the light source (5) to the needle section is 5 cm to 40 cm.

17. Method for monitoring needles (3) during operation of a knitting machine according to one of claims 10 to 16, characterized in that the light source (5) is used to monitor the light emitted by the needles (3) incident light rays (6) are emitted as a convergent beam.

18. Method for monitoring needles (3) during operation of a knitting machine according to one of claims 10 to 17, characterized in that the control unit (11) determines a median value from the image data of several images for the distances between the imaged needles (3) and a moving average value as a standard distance from the median values ​​of the images, and in order to determine a faulty or missing needle (3), each individual distance determined between two needles (3) is compared with the standard distance.