Adjustment method for automatic knitting machine for plaiting knitting and corresponding system
By using a digital camera and automatic image recognition technology in the yarn-adding knitting machine, the position of the yarn guide is automatically adjusted, solving the problems of slow adjustment and excessive waste caused by reliance on experience in the existing technology. This achieves efficient and reliable yarn guide adjustment and improved yarn-adding quality.
Patent Information
- Application Number
- CN202480044688.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Priority Date
- 2023-06-07
- Filing Date
- 2024-06-05
- Publication Date
- 2026-01-27
AI Technical Summary
In existing yarn-adding knitting machines, the adjustment of the yarn guide position relies on the worker's experience, resulting in a slow, cumbersome adjustment process that generates a lot of waste and makes it difficult to ensure the quality of yarn addition.
A digital camera is used to capture images of straight segments of the yarn. Automatic image recognition technology is used to determine the yarn direction, and the position of the guide is adjusted automatically or manually to ensure that the yarn is correctly guided to the knitting needle.
It enables faster and more reliable yarn guide adjustment, reduces or avoids waste, and improves yarn feeding quality.
Smart Images

Figure CN121420102A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of automatic knitting machines for plating knitting.
[0002] More specifically, the present invention relates to an adjustment method for an automatic knitting machine for yarn-adding knitting, the automatic knitting machine (1) comprising: Multiple movable knitting needles; At least one yarn delivery device that delivers a first yarn and a second yarn to the same working needle, said working needle being one of said movable needles, which captures the first yarn and the second yarn at a specific moment, said yarn delivery device including a first guide for the first yarn and a second guide for the second yarn, the first yarn extending from the first guide along a first straight segment in a first direction to the working needle, and the second yarn extending from the second guide along a second straight segment in a second direction different from the first direction to the working needle, and An actuation device for automatically moving the knitting needles according to a predetermined pattern to manufacture a padded knitted fabric using the first and second yarns conveyed by the yarn conveying device.
[0003] The terms "plaiting" / "plaited" used in the textile industry have the same meaning as the terms "plating" / "plated" used in this article.
[0004] Added knit fabric is a fabric in which the loops consist of two (or more) loops of yarn of different types or colors, one yarn protruding on one side of the fabric and the other on the opposite side. In automatic knitting machines used for added knitting, two (or more) yarns in different positions are supplied to the same needle separately by means of their own guides. Depending on the position of the yarn within the needle hook, each yarn will emerge from one side or the other of the knit fabric. Background Technology
[0005] Defects will appear on yarn-stitched fabrics when the yarn is on the wrong side of the fabric. The main reason is the incorrect positioning of the guides that supply the yarn to the needles. The correct position of the guides is not easy to determine because it depends on several factors, including the mechanical properties of each yarn and the working tension range of the yarn. In practice, in automatic knitting machines used for yarn-stitching, the general method for adjusting the position of the guides is based on trial and error using the experience of a skilled worker. The worker starts the machine to produce the first section of fabric, stops the machine, and visually inspects the produced section of fabric. If he / she finds any yarn-stitching defects, the relative position of the yarn guides is corrected. Because this correction is based on the worker's experience and knowledge, it usually requires repeated trial and error to achieve good results. When this process is finished, the produced fabric is waste and must be discarded. On the other hand, it must be considered that, taking a typical yarn-stitching circular knitting machine as an example, it may have up to 92 conveyor devices for feeding two (or more) yarns, each equipped with two (or more) yarn guides, and the relative position of the yarn guides must be adjusted using this iterative trial and error method. Therefore, the adjustment process to ensure the quality of yarn addition is slow and tedious, generates a large amount of waste fabric, and requires the intervention of experienced workers.
[0006] WO2022158251A1 discloses a method for automatically adjusting a linear knitting machine, particularly suitable for manufacturing fabrics with reverse-insertion yarns, where the positions of two yarns are reversed at predetermined locations to obtain a pattern on both sides of the fabric. In this case, the specific objective of the method is to precisely control the position of the reversal in the yarn-insertion fabric. As in the aforementioned prior art, this is also a trial-and-error method based on inspecting already produced fabrics, but it focuses on controlling the reversal position in the yarn-insertion fabric and is executed automatically by an AI-based system that uses a camera to inspect the yarn-insertion fabric being produced. The system uses the camera to inspect the yarn-insertion fabric being produced, determines the reversal position in the yarn-insertion fabric, and adjusts a set of parameters, including the relative positions of guides, when the position differs from the expected position. Summary of the Invention
[0007] The object of this invention is to provide a method for adjusting an automatic knitting machine of the type described in the "Technical Field" section above for yarn-feeding knitting, which allows for adjustments in a more convenient and reliable manner and produces less waste or no waste at all.
[0008] This is achieved by adjusting an automatic knitting machine used for yarn-feeding knitting, the key design point of which is that a digital camera is arranged to capture image frames of an area including at least one of the following: at least a portion of a first straight segment of a first yarn and at least a portion of a second straight segment of a second yarn. The method includes the following steps: [a] Obtain an image frame from the digital camera that includes at least one of the following: at least a portion of the first straight line segment and at least a portion of the second straight line segment; [b] Perform automatic image recognition in at least one of the image frames to identify at least one of the following: the portion of the first straight line segment and the portion of the second straight line segment; [c] From the automatic image recognition performed in step [b], at least one of the following is automatically derived: the first direction of the first line segment and the second direction of the second line segment; [d] Based on the first or second direction automatically derived in step [c], adjust at least one of the following: the relative position of the first guide relative to the working needle and the relative position of the second guide relative to the working needle.
[0009] Most commonly, automatic knitting machines for yarn-padded knitting are suitable for producing yarn-padded fabrics with two different yarns. In this case, each yarn conveying device has two guides that feed the two different yarns (the first yarn and the second yarn) to the same working needle. The invention is not limited to this double-yarn structure. The invention is also applicable to automatic knitting machines for yarn-padded knitting, which are suitable for producing yarn-padded fabrics with more than two different yarns (usually three yarns), wherein each yarn conveying device has more than two guides that can feed more than two yarns to the same working needle. For example, if there are three yarns and three guides, the area of the image frame captured by the digital camera also includes the third straight segment of the third yarn, which is automatically identified in step [b], the third direction of the third straight segment is automatically derived in step [c], and in step [d], the relative position of the third guide can also be adjusted according to the third direction.
[0010] In some knitting machines, it is feasible to install a digital camera on each conveyor. However, this would significantly increase the space requirement and complicate access to necessary machine parts (needles, the conveyor itself, etc.). Therefore, in a preferred embodiment, a single digital camera or a small number of digital cameras are used, moved from one conveyor to another to apply this method. For example, the digital cameras can be easily moved along tracks provided for this purpose.
[0011] Preferably, the first and second directions are defined by the angle between each direction and a reference direction. The reference direction can be a fixed direction, such as a horizontal direction. It can also be a direction associated with a part of the knitting machine, preferably associated with the working needle or a weaving needle (including the working needle). For example, as will be discussed later, it can be the direction of a local path followed by a weaving needle, including the working needle.
[0012] The yarn conveying device can be a single integral unit, in which a first guide and a second guide are formed, or the yarn conveying device can be two independent devices, each supporting the first guide and the second guide respectively.
[0013] As will be seen in the detailed description of the embodiments below, the method according to the invention can be easily implemented simply by installing a digital camera in front of the area where the yarn feeder feeds two or more yarns to the working needles. Installing the camera in this location generally poses no problem. The camera can be attached to the knitting machine or an external frame. The main advantage of the invention is that it allows the automated system or the worker (not necessarily an experienced worker in adjusting the knitting machine used for yarn feeding) to make adjustments only by focusing on simple parameters automatically provided by the method. In fact, to ensure correct yarn feeding, it is sufficient to know only the direction of the first straight segment (first direction) and / or the direction of the second straight segment (second direction), which are directly obtained by the digital camera and automatic recognition, and these directions are corrected by adjusting the relative positions of the first guide and / or the second guide. In this case, the use of a digital camera and automatic recognition software is very effective because straight yarns are particularly easy to capture in the image and can be automatically identified and characterized with very high reliability.
[0014] The method is more robust when the directions of the first and second straight segments are obtained and used to adjust the relative positions of the first and second guides. In a preferred embodiment, the digital camera is configured to capture image frames comprising at least a portion of the first and second straight segments; and: In step [a], the image frame obtained from the digital camera simultaneously contains at least a portion of the first straight line segment and at least a portion of the second straight line segment; In step [b], at least a portion of the first straight line segment and at least a portion of the second straight line segment are identified by automatic image recognition; In step [c], the first and second directions are automatically derived; In step [d], at least one of the relative positions of the first guide relative to the working needle and the second guide relative to the working needle is adjusted based on both the first direction and the second direction automatically derived in step [c].
[0015] In a preferred embodiment, the adjustment in step [d] is based on at least one (preferably both) of the following: a comparison between a first direction and a predetermined target direction of the first direction, and a comparison between a second direction and a predetermined target direction of the second direction. These predetermined target directions can be obtained from a database or derived from a parametric function formula, the parameters of which include, for example, yarn type, yarn tension, needle type, etc. For a particular knitting machine, a reliable database or formula can be obtained from previous tests. When this is achieved, it is not necessary to inspect the fabric, and therefore the method can be performed with the knitting machine stopped or for a very short period of time, resulting in little or no waste. When the automatic knitting machine is properly adjusted, the first and second directions obtained by performing steps [a] to [c] of this method, and these directions are considered to be the optimal directions for achieving the correct yarn feeding effect, can also be directly used as predetermined target directions for the first and second directions.
[0016] In some embodiments, the adjustment in step [d] is performed automatically.
[0017] In other embodiments, between steps [c] and [d], instructions regarding the adjustment are automatically generated and displayed on the screen, and in step [d], the adjustment is performed manually by the worker.
[0018] In some embodiments, a holographic projector is arranged to project a hologram in the area (the area where the image frame is captured by the digital camera), and between steps [c] and [d], the holographic projector projects at least one of the following in the area: a hologram of a first straight line in a predetermined target direction of the first direction, and a hologram of a second straight line in a predetermined target direction of the second direction. As described above, these predetermined target directions can be obtained from a database or from a formula, or they can be the first and second directions obtained by performing steps [a] to [c] of this method when it is known that the automatic knitting machine is well adjusted for correct yarn feeding. The function of these holograms is to provide a visual reference for the worker when he / she manually makes the adjustments in step [d]. The worker can adjust the relative positions of the first guide and / or the second guide until the first straight line segment and / or the second straight line segment match the hologram of the first or second straight line, respectively.
[0019] In a preferred embodiment, the automatic knitting machine is configured such that, when it is in operation, the knitting needles move cyclically one after another along a path, and the digital camera is arranged such that the area in which the digital camera captures an image frame includes a set of said knitting needles (including working needles), which move cyclically one after another along a local path in said area; and: Step [b] includes performing automatic image recognition in at least one of the image frames to identify at least a portion of the local path adjacent to and upstream of the working needle (the term "upstream" is relative to the direction of movement of the needle). Step [c] includes automatically deriving the slope of the portion of the local path, the slope defining a third direction; Furthermore, in step [c], at least one of the following is automatically derived from the automatic image recognition performed in step [b]: the angle between the first direction and the third direction, and the angle between the second direction and the third direction.
[0020] Adjusting based on these angles is particularly effective: it can reliably ensure the production of defect-free, properly fitted yarn.
[0021] Preferably, the portion of the path seen by the digital camera is a straight line, such that the third direction is the direction of the straight line. For example, as will be shown in the detailed description of the embodiments later, for a circular knitting machine, the digital camera can be positioned such that it focuses on a knitting needle that appears to the digital camera to be moving cyclically along a straight line. This makes the method more robust because the straight line can be more reliably and automatically identified and characterized.
[0022] In some embodiments, the digital camera captures image frames as the knitting needles move cyclically along the path one after another, and in step [b], the automatic image recognition is performed by superimposing multiple temporally consecutive image frames, in which a continuous line along a portion of a local path is formed by a segment of at least some knitting needles moving cyclically along that portion of the local path, and in step [b], automatic recognition of the local path portion is achieved by automatically recognizing the continuous line. In these embodiments, the knitting machine must be in operation when the image frames are captured. However, since there is no need to inspect the produced fabric, the knitting machine must run for a very short time. Optionally, the adjustment in step [d] can be done automatically. In this case, the method can be applied continuously while the knitting machine is running.
[0023] Preferably, the continuous line is formed by spots in the image frame, each spot corresponding to light reflection in the segment of each knitting needle. The continuous line generated by moving the spots can be automatically identified, and its characteristics can be determined with greater reliability.
[0024] In other embodiments, in step [b], automatic image recognition is performed in one of the image frames, and a pattern defined by multiple segments of at least some knitting needles is automatically identified, and said portion of the local path is automatically identified as a line passing through said segment.
[0025] Preferably, the pattern comprises multiple spots in the image frame, each spot corresponding to light reflection in a segment of each needle. This provides a particularly quick and simple method for determining the pattern, as a set of spots is easily identified by the software. This solution is particularly suitable for knitting machine needles, which are typically metallic, have surfaces capable of reflecting light well, and possess various curved portions that can generate spots reflecting light in the image frame. Furthermore, the method according to this solution can be easily transferred from one knitting machine to another with only minor adjustments, because even if the needle shapes are different, the patterns of a set of spots generated by equivalent portions of the needles can be very similar.
[0026] Preferably, the section of the knitting needle is the free end of the needle. This is a portion of the needle that is easier to capture an image of and often forms a hook, thus producing good contrast in the image frame.
[0027] Preferably, the controlled illumination device is focused on the area where the digital camera captures the image frame, so that the image to be identified is independent of the ambient light at the location of the machine.
[0028] Although the method according to the invention can be applied to various automatic knitting machines, in a preferred embodiment, the automatic knitting machine is a circular knitting machine, wherein the knitting needles are arranged in a rotating cylinder such that the knitting needles travel along a circumference coaxial with the rotating cylinder. The digital camera 7 is statically positioned such that it does not rotate with the rotating cylinder, and the area in which the digital camera captures image frames includes a portion of the circumference.
[0029] The invention also includes a system for adjusting an automatic knitting machine used for yarn-feeding knitting, the system comprising an automatic knitting machine for yarn-feeding knitting, the automatic knitting machine having: Multiple movable knitting needles; At least one yarn delivery device that delivers a first yarn and a second yarn to the same working needle, said working needle being one of said movable needles, which captures the first yarn and the second yarn at a specific moment, said yarn delivery device including a first guide for the first yarn and a second guide for the second yarn, the first yarn extending from the first guide along a first straight segment in a first direction to the working needle, and the second yarn extending from the second guide along a second straight segment in a second direction different from the first direction to the working needle, and An actuation device for automatically moving the knitting needles according to a predetermined pattern to manufacture a padded knitted fabric using the first and second yarns conveyed by the yarn conveying device. Its key design features include: - A digital camera, which is arranged to capture an image frame comprising at least one of the following: at least a portion of a first straight line segment and at least a portion of a second straight line segment; - A processor, which is connected to the digital camera; - A computer program including instructions that, when executed by the processor, cause the processor to perform the following steps: [a] Obtain an image frame from the digital camera that includes at least one of the following: at least a portion of the first straight line segment and at least a portion of the second straight line segment; [b] Perform automatic image recognition in at least one of the image frames to identify at least one of the following: the portion of the first straight line segment and the portion of the second straight line segment; [c] From the automatic image recognition performed in step [b], at least one of the following is automatically derived: the first direction of the first line segment and the second direction of the second line segment; [d] Based on the first or second direction automatically derived in step [c], adjust at least one of the following: the relative position of the first guide relative to the working needle and the relative position of the second guide relative to the working needle; or, provide instructions or guidance for performing the adjustment based on the first or second direction.
[0030] The system may optionally have structural features according to a preferred embodiment of the method described above, and the computer program may optionally include instructions for performing the steps of the method according to the preferred embodiment.
[0031] The present invention also includes the computer programs defined above in the system description.
[0032] The invention also includes other features relating to the details shown in the detailed description and accompanying drawings of embodiments of the invention.
[0033] The invention also includes other features relating to the details shown in the detailed description and accompanying drawings of embodiments of the invention. Attached Figure Description
[0034] The advantages and features of the present invention will become apparent from the following description, in which preferred embodiments are described with reference to the accompanying drawings, which do not limit the scope of the main claims.
[0035] Figure 1 This is a schematic diagram of the system according to the present invention. The automatic knitting machine is a circular knitting machine, having a set of vertical needles and a set of sinkers.
[0036] Figure 2 This is a top view of the knitting needle, with the latch in the open position.
[0037] Figure 3 This is a view of the knitting needle from the same top, with the needle tab in the closed position.
[0038] Figure 4 This is a view of the area captured by the digital camera, as seen from the digital camera itself.
[0039] Figure 5 It was taken with a digital camera. Figure 4 Image frames of the region.
[0040] Figure 6 This is a schematic diagram of a processed image for automatic image recognition according to the first embodiment.
[0041] Figure 7 yes Figure 6 The processed image is a real photograph.
[0042] Figure 8 yes Figure 7 A magnified view of a portion of the image.
[0043] Figure 9 It shows from Figure 8 Exported direction and angle.
[0044] Figure 10 yes Figure 4 A magnified view of a portion of the image.
[0045] Figure 11 This is a schematic diagram of a processed image for automatic image recognition according to the second embodiment, and corresponds to... Figure 10 The enlarged area shown in the image.
[0046] Figure 12 It shows from Figure 11 Exported direction and angle.
[0047] Figure 13 This is a block diagram of the main steps of the method according to the present invention. Detailed Implementation
[0048] Figures 1-9 A first embodiment of the system and method according to the present invention. Figures 10-12 The second embodiment differs from the first embodiment only in how some steps of the method are performed. The automatic knitting machine 1 for yarn-feeding is the same in both embodiments. It is a circular knitting machine with a set of vertical needles 2 and a set of sinkers 16 for producing yarn-feeding fabric 17.
[0049] Figure 1This is a schematic diagram of the system, in which an automatic knitting machine 1 for yarn feeding and knitting is shown in a schematic cross-sectional view. The system includes the automatic knitting machine 1, a digital camera 7, a controlled lighting device 10, and a computer system including a processor 12 connected to the digital camera 7. A computer program, including software for automatic image recognition, is executed by the processor 12.
[0050] The automatic knitting machine 1 includes multiple yarn feed devices 3, multiple movable needles 2, and an actuation device. As the needles 2 move, each needle 2 is adapted to subsequently catch and release two yarns 4a, 4b supplied by one of the yarn feed devices 3. The actuation device is used to automatically move the needles 2 according to a predetermined pattern to produce a layered knitted fabric using the two yarns 4a, 4a supplied by the yarn feed devices 3.
[0051] Each yarn feeder 3 feeds the first yarn 4a and the second yarn 4b to the same working needle 22. The term "working needle" refers to one of the movable needles 2 that simultaneously catches the first yarn 4a and the second yarn 4b at a given moment. In the embodiments described herein, as... Figure 4 As shown, each yarn conveying device 3 is a single arm, including a first guide 5a for a first yarn 4a and a second guide 5b for a second yarn 4b. The first guide 5a is a hole in the arm through which the first yarn 4a passes. The second guide 5b is a groove formed at the lower end of the arm, which guides the second yarn 4b. Other embodiments are also possible, wherein the conveying device 3 consists of two separate components, each including a first guide 5a and a second guide 5b. Figure 4 As shown, a first yarn 4a extends from a first guide member 5a along a first straight segment 6a in a first direction D1 to a working needle 22. A second yarn 4b extends from a second guide member 5b along a second straight segment 6b in a second direction D2, which is different from the first direction D1. Each yarn feed device 3 receives two yarns 4a and 4b from yarn spools 15a and 15b.
[0052] Both the method and the system were tested using an automatic circular knitting machine for yarn-feeding knitting (i.e., a CANMARTEX-JUMBERCA model with 1728 needles, a diameter of 30 inches, and a 18-needle pitch). The needles 2 are arranged in a rotating cylinder 11 such that they travel along a circumferential path coaxial with the cylinder 11. The needles 2 interact with a cam, which is stationary in a dial about the rotating cylinder 11. As the cylinder 11 rotates continuously, the cam causes each needle 2 to move vertically up and down, forming a knitted loop with two yarns 4a and 4b. The motor-driven rotating cylinder 11 and the stationary cam are the actuation devices described above for the automatic movement of the needles 2. Because the cam is stationary, each needle 2 traveling along the circumference has a unique position at each point on the circumference. A sinker 16 is arranged between the needles 2 and moves horizontally during machine knitting.
[0053] All two knitting needles are identical. They are latch needles, such as... Figure 2 and Figure 3 As shown. Each knitting needle 2 has a free tip forming a hook 13 and a pivoting tongue 14, which is pushed by the yarns 4a and 4b of the knitted loops as the knitting needle 2 moves up and down, causing the tongue 14 to subsequently close and open the hook 13. As the knitting needle 2 moves, the hook 13 is adapted to subsequently grasp and release the two yarns 4a and 4b. Figure 2 and Figure 3 The knitting needle 2 and the needle latch 14 are shown in the open and closed positions, respectively. The operation of this circular knitting machine with knitting needles and sinkers, as well as the movement of the needles with locking latches forming knitting loops, will not be described in detail here, as they are well known to those skilled in the art.
[0054] The digital camera 7 is statically set up, so it does not rotate with the rotating cylinder 11. It is arranged to capture image frames including a portion of a first straight segment 6a, a portion of a second straight segment 6b, and a weaving needle 2 (including a working needle 22). Figure 4 This is a diagram of the area seen in an image frame taken with a digital camera 7. Figure 5 It was taken with a digital camera 7. Figure 4 The image frame of the region. This region is a portion of the circumference along which the knitting needle 2 travels. Because this portion is a small segment of the circumference, and the digital camera 7 is focused perpendicular to the circumferential axis, in the digital frame captured by the digital camera 7, the knitting needle appears to move cyclically along the local path 8 one after another, as shown. Figure 6 As illustrated, the path is essentially a straight line along a third direction, D3.
[0055] The illumination device 10 is a statically positioned lamp whose light is focused on the area. The relative position of the lamp to the digital camera 7 and the intensity of the light emitted by the lamp are adjusted to obtain suitable light reflection on the knitting needle 2 and on the first and second straight segments 6a of the two yarns 4a, 4b, allowing the pattern to be identified in the processed image frame (5) as described below. In the exemplary embodiment shown in the figures, the digital camera 7 and the illumination device 10 are fixed to a support on the outside of a virtual cylinder relative to the axially extending rotating needle cylinder (11). Optionally, the support may be a circular channel along which the digital camera 7 can move from one area to another, each of which corresponds to a knitting needle 2 supplied by one of the yarn delivery devices 3. In other possible embodiments, the digital camera 7 may also be a handheld portable camera, such as a handheld scanner. The digital camera 7 and / or the illumination device 10 may also be arranged in different positions. For example, they may be arranged on the inside relative to the virtual cylinder.
[0056] The digital camera used in the following tests is a TIS-DMK-33UX264 video camera sold by The Imaging Source Europe GmbH, a German company. It is equipped with a Sony IMX264 sensor, featuring a resolution of 2448x2048 pixels and a video recording speed of 35 PFS (frames per second). This sensor employs a global shutter CMOS image capture method, allowing all data to be collected simultaneously without shutter lag.
[0057] The method according to the present invention includes the following main steps, which are automatically executed by a computer program executed by processor 12: [a] The above reference was obtained from a digital camera 7. Figure 4 and Figure 5 The image frame of the described area includes a portion of a first straight segment 6a, a portion of a second straight segment 6b, and a tissue needle 2 including a working knitting needle 22; [b] Perform automatic image recognition in at least one of the image frames to identify a portion of the first straight line segment 6a, a portion of the second straight line segment 6b, and a portion of the local path 8 of the needle 2 adjacent to and upstream of the working needle 22; [c] Based on the automatic image recognition performed in step [b], the first direction D1 of the first straight line segment 6a, the second direction D2 of the second straight line segment 6b, and the third direction D3 of the local straight path 8 are automatically derived. The angle A1 between the first direction D1 and the third direction D3, as well as the angle between the second direction D2 and the third direction D3, are also automatically derived.
[0058] The final step [d] of the method can be performed manually or automatically, including adjusting the relative position of the first guide 5a relative to the working needle 22 and / or the relative position of the second guide 5b relative to the working needle 22 according to the first direction D1 and / or the second direction D2 automatically derived in step [c]. Preferably, the adjustment is made according to angles A1 and / or A2.
[0059] exist Figures 6-9 In the illustrated embodiment, as the automatic knitting machine 1 operates and the knitting needles 2 move cyclically along the path one after another, the digital camera 7 captures image frames. In step [b], automatic image recognition is performed in the superposition of a plurality of temporally consecutive image frames. This superposition can be the sum of a series of image frames captured by the digital camera 7, or equivalently, it can be an overexposed image captured by the digital camera 7. Because the knitting needles 2 move cyclically along a straight local path 8, a continuous straight line 18 is formed along a portion of the local path 8 in this superposition. More specifically, in Figures 6-9 In the illustrated embodiment, the line 18 is formed by spots 9 in the image frame, each spot 9 corresponding to the light reflection from the free end of each knitting needle 2. These spots 9 are... Figure 4 and Figure 5 As shown in the diagram. They cycle along a straight local path 8, thus forming line 18 in the superposition, as... Figure 6 and Figure 7 As shown. In step [b], the portion of the local path 8 is automatically identified by automatically recognizing the continuous lines 18 in the overlay. Computer program analysis is as follows. Figure 8 The superimposed magnified view shows a portion of the first straight line segment 6a. Figure 8 (lower line in the middle), part of the second straight line segment 6b ( Figure 8 The middle line in the middle) and part of the local path 8 of the straight line ( Figure 8 (The upper limit of the recognition), and from this recognition, three directions D1, D2, D3 and two angles A1, A2 are derived, such as Figure 9 As shown.
[0060] The two angles A1 and A2 are automatically compared with predetermined values. If a significant difference is found, the relative positions of the first guide 5a and the second guide 5b are adjusted. This can be achieved by adjusting the position of the arm including the two guides 5a and 5b, or by adjusting the position of the needle 2, so that angles A1 and A2 match the predetermined values. If the automatic knitting machine 1 includes a device for moving the arm or needle 2 in a controlled manner, the adjustment in step [d] can be performed automatically and can be controlled by a computer program. However, in most cases, the machine 1 does not have these devices, and the adjustment in step [d] is performed manually by the worker. Instructions for the adjustments to be made can be automatically generated by the computer program and displayed on the screen, allowing the worker to follow these instructions to make adjustments in step [d]. For example, these instructions may include the distance the arm must move in the vertical direction.
[0061] Another possibility for assisting workers in making adjustments (not shown in the figure) lies in using a holographic projector configured to project a hologram in the area. Between steps [c] and [d], the computer program automatically causes the holographic projector to project a hologram of a first straight line in a predetermined target direction of a first direction D1 and a hologram of a second straight line in a predetermined target direction of a second direction D2 in the area. Preferably, when it is known that the automatic knitting machine is well adjusted for correct yarn feeding, these predetermined target directions can be the first direction D1 and the second direction D2 initially obtained by performing steps [a] to [c] of this method.
[0062] Figure 10-12 The illustrated embodiment can be implemented while the automatic knitting machine 1 is operating or stopped. It differs from the previously discussed embodiments in that, in step [b], automatic image recognition is performed in one of the image frames captured by the digital camera 7, wherein a pattern defined by multiple segments of some knitting needles 2 is automatically identified. In the exemplary embodiment shown in the figures, the pattern is the aforementioned plurality of spots 9. Parts of the local path 8 are automatically identified as straight lines that better pass through these spots 9. Computer program analysis is as follows... Figure 11 The processed image shown identifies a portion of the first straight line segment 6a. Figure 12 (the lower line in the middle) and part of the second straight line segment 6b ( Figure 11 It identifies the upper line of the pattern (the line through which the pattern passes), and derives two directions D1 and D2 from this identification. It also identifies the pattern formed by the arranged dots 9, and derives a third direction D3 from this identification, which is the direction of the straight line passing through dots 9. (See also...) Figure 12 As shown, two angles A1 and A2 are then derived from the three directions D1, D2, and D3. Figure 11 Image processing is done by, for example Figure 10 The image shown is a magnified partial view of an image frame taken by a digital camera 7, which is oversaturated and converted to monochrome.
Claims
1. A method for adjusting an automatic knitting machine to perform yarn-feeding knitting, said automatic knitting machine (1) comprising: Multiple movable knitting needles (2); At least one yarn delivery device (3) delivers a first yarn (4a) and a second yarn (4b) to the same working needle (22), the working needle (22) being one of the movable needles (2) that captures the first yarn (4a) and the second yarn (4b) at a specific moment. The yarn delivery device (3) includes a first guide (5a) for the first yarn (4a) and a second guide (5b) for the second yarn (4b). The first yarn (4a) extends from the first guide (5a) along a first straight segment (6a) in a first direction (D1) to the working needle (22), and the second yarn (4b) extends from the second guide (5b) along a second straight segment (6b) in a second direction (D2) different from the first direction (D1) to the working needle (22). An actuation device is used to automatically move the knitting needle (2) according to a predetermined pattern in order to manufacture a padded knitted fabric using the first yarn (4a) and the second yarn (4b) conveyed by the yarn conveying device (3); The digital camera (7) is characterized in that it is arranged to capture image frames comprising at least one of the following regions: at least a portion of the first straight line segment (6a) and at least a portion of the second straight line segment (6b); and The method includes the following steps: [a] Obtain an image frame from the digital camera (7) comprising at least one of the following: at least a portion of the first straight line segment (6a) and at least a portion of the second straight line segment (6b); [b] Perform automatic image recognition in at least one of the image frames to identify at least one of the following: the portion of the first straight line segment (6a) and the portion of the second straight line segment (6b); [c] From the automatic image recognition performed in step [b], at least one of the following is automatically derived: the first direction (D1) of the first straight line segment (6a) and the second direction (D2) of the second straight line segment (6b); [d] Based on the first direction (D1) or the second direction (D2) automatically derived in step [c], adjust at least one of the following: the relative position of the first guide (5a) relative to the working needle (22) and the relative position of the second guide (5b) relative to the working needle (22).
2. The method according to claim 1, wherein, The digital camera (7) is configured to capture image frames comprising at least a portion of the first straight line segment (6a) and at least a portion of the second straight line segment (6b); and In step [a], the image frame obtained from the digital camera (7) comprises at least a portion of the first straight line segment (6a) and at least a portion of the second straight line segment (6b); In step [b], at least a portion of the first straight line segment (6a) and at least a portion of the second straight line segment (6b) are identified by automatic image recognition. In step [c], the first direction (D1) and the second direction (D2) are automatically derived; In step [d], based on the first direction (D1) and the second direction (D2) automatically derived in step [c], at least one of the following is adjusted accordingly: the relative position of the first guide (5a) relative to the working needle (22) and the relative position of the second guide (5b) relative to the working needle (22).
3. The method according to any one of claims 1 to 2, wherein, In step [d], the adjustment of at least one of the relative positions of the first guide (5a) relative to the working needle (22) and the relative positions of the second guide (5b) relative to the working needle (22) is performed automatically.
4. The method according to any one of claims 1 to 3, wherein, Between steps [c] and [d], instructions for adjusting at least one of the relative positions of the first guide (5a) relative to the working needle (22) and the second guide (5b) relative to the working needle (22) are automatically generated and displayed on the screen, and in step [d], the adjustment is performed manually.
5. The method according to any one of claims 1 to 3, wherein, A hologram projector is arranged to project a hologram in the region, wherein between steps [c] and [d], the hologram projector projects at least one of the following in the region: a hologram of a first straight line in a predetermined target direction of the first direction (D1) and a hologram of a second straight line in a predetermined target direction of the second direction (D2).
6. The method according to any one of claims 1 to 5, wherein, The automatic knitting machine (1) is configured such that when it is in operation, the knitting needles (2) move cyclically one after another along a path, and the digital camera (7) is arranged such that the area in which the digital camera (7) captures an image frame includes a set of knitting needles (2) moving cyclically one after another along a local path (8) in the area, including the working knitting needles (22). and: Step [b] includes performing automatic image recognition in at least one of the image frames to identify at least a portion of the local path (8) that is adjacent to and upstream of the working needle (22); Step [c] includes automatically deriving the slope of said portion of the local path (8), the slope defining a third direction (D3); Furthermore, in step [c], at least one of the following is automatically derived from the automatic image recognition performed in step [b]: the angle (A1) between the first direction (D1) and the third direction (D3) and the angle (A2) between the second direction (D2) and the third direction (D3).
7. The method according to claim 6, wherein, The portion of the local path (8) is a straight line as seen from the digital camera (7), such that the third direction (D3) is the direction of the straight line.
8. The method according to any one of claims 6 or 7, wherein, The digital camera (7) captures the image frames as the knitting needles (2) move cyclically along the path one after another, and in step [b], the automatic image recognition is performed by superimposing multiple image frames that are sequentially consecutive in time, wherein, in the superimposition, a continuous line (18) along the portion of the local path (8) is formed by at least some segments of the knitting needles (2) that move cyclically along the portion of the local path (8). Furthermore, in step [b], the automatic identification of the portion of the local path (8) is achieved by automatically identifying the continuous line (18).
9. The method according to claim 8, wherein, The continuous line (18) is formed by spots (9) in the image frame, each spot (9) corresponding to light reflection in the segment of each needle (2).
10. The method according to any one of claims 6 or 7, wherein, In step [b], the automatic image recognition is performed in one of the image frames, and a pattern defined by multiple segments of at least some knitting needles (2) is automatically recognized, and the portion of the local path (8) is automatically recognized as a line passing through the segment.
11. The method according to claim 10, wherein, The pattern comprises a plurality of spots (9) in the image frame, each spot (9) corresponding to light reflection in the segment of each needle (2).
12. The method according to any one of claim 10 or 11, wherein, The section of the knitting needle (2) is the free end of the knitting needle (2).
13. The method according to any one of claims 1 to 12, wherein, The controlled lighting device (10) is focused on the area where the digital camera (7) captures the image frame.
14. The method according to any one of claims 1 to 13, wherein, The automatic knitting machine (1) is a circular knitting machine, wherein the knitting needles (2) are arranged in a rotating cylinder (11) that causes the knitting needles (2) to travel along a circumference coaxial with the rotating cylinder (11), and wherein the digital camera (7) is stationary such that it does not rotate with the rotating cylinder (11), and the area of the image frame captured by the digital camera (7) includes a portion of the circumference.
15. A system for adjusting an automatic knitting machine for yarn-feeding knitting, comprising an automatic knitting machine (1) for yarn-feeding knitting, said automatic knitting machine having: Multiple movable knitting needles (2); At least one yarn delivery device (3) delivers a first yarn (4a) and a second yarn (4b) to the same working needle (22), the working needle (22) being one of the movable needles (2) that captures the first yarn (4a) and the second yarn (4b) at a specific moment. The yarn delivery device (3) includes a first guide (5a) for the first yarn (4a) and a second guide (5b) for the second yarn (4b). The first yarn (4a) extends from the first guide (5a) along a first straight segment (6a) in a first direction to the working needle (22), and the second yarn (4b) extends from the second guide (5b) along a second straight segment (6b) in a second direction different from the first direction to the working needle (22). An actuation device is used to automatically move the knitting needle (2) according to a predetermined pattern in order to manufacture a padded knitted fabric using the first yarn (4a) and the second yarn (4b) conveyed by the yarn conveying device (3); Its features are, It also includes: - A digital camera (7) arranged to capture an image frame comprising at least one of the following: at least a portion of the first straight line segment (6a) and at least a portion of the second straight line segment (6b); - Processor (12) connected to the digital camera (7); - A computer program including instructions that, when executed by the processor (12), cause the processor (12) to perform the following steps: [a] Obtain an image frame from the digital camera (7) comprising at least one of the following: at least a portion of the first straight line segment (6a) and at least a portion of the second straight line segment (6b); [b] Perform automatic image recognition in at least one of the image frames to identify at least one of the following: the portion of the first straight line segment (6a) and the portion of the second straight line segment (6b); [c] From the automatic image recognition performed in step [b], at least one of the following is automatically derived: the first direction (D1) of the first straight line segment (6a) and the second direction (D2) of the second straight line segment (6b); [d] Based on the first direction (D1) and the second direction (D2) automatically derived in step [c], adjust at least one of the following: the relative position of the first guide (5a) relative to the working needle (22) and the relative position of the second guide (5b) relative to the working needle (22); or, provide instructions or guidance for performing the adjustment based on the first direction (D1) or the second direction (D2).
Citation Information
Patent Citations
Production method and production system for correction data for inverse plating
WO2022158251A1