Methods and systems for line-by-line inspection of one-dimensional fabrics

Abstract

Systems and methods for inspecting fabrics on a loom include: an imaging device that collects images of the weaving area of ​​the loom; and a frame acquisition unit that receives images of the weft yarns and sends compressed image data packets to an image processor. Irregularities can be detected by comparing a string of numbers representing a characteristic sequence of warp points and warp sinkers along the weft yarns with the corresponding rows (901) of a reference matrix (900) representing the desired warp points and desired warp sinkers of the desired weave pattern. The string of numbers can be a sequence of Boolean values.

Description

Technical Field

[0001] This invention relates to systems and methods for inspecting fabrics on looms. In particular, this invention relates to row-by-row weft inspection. Background Technology

[0002] Weaving is the most popular method of fabric manufacturing. It is primarily accomplished by interlacing two sets of orthogonal yarns (warp and weft) into regular and cyclical patterns. Weaving involves the sequential repetition of shedding, shuttle insertion, and beat-up operations. All these processes are typically performed on a loom. Shedding is the process of raising or lowering the warp yarns to create a space called the shed, through which the weft yarns can pass. Shuttle insertion is the process of inserting the weft yarn through the shed, causing it to intersect with the warp yarns. Beat-up is the process of pressing the weft yarn against the shed, forming a new woven fabric at the shed.

[0003] Fabric defects can occur during the weaving process. These defects include broken yarns, double yarns, holes, loose yarns, and stains. The quality of a woven fabric depends on the number of defects remaining in the fabric after the manufacturing process. Defects generated during any of these processes determine the quality of the finished fabric. Typically, finished fabric defects are inspected according to industry standards and graded using quality indices. For example, in a standard four-point system for fabric inspection, deductions are assigned to detected defects. The number of deductions also depends on the length of the defect: 1 point for defects 3 inches or less, 2 points for defects between 3 and 6 inches, 3 points for defects between 6 and 9 inches, and 4 points for defects longer than 9 inches. The quality of this batch of fabric is described by the deductions per 100 yards of inspected fabric, with a maximum defect rate of 40 points generally considered acceptable. Besides the four-point system, other standard indices, such as the more complex ten-point system for knitted fabrics or the Dallas System, can also be used to measure fabric quality.

[0004] Finished fabrics are typically inspected manually. During manual inspection, the sample size usually checked is at least ten percent of a roll of finished fabric. Defects in uninspected rolls usually go undetected until the fabric is sold. Furthermore, although this defect inspection is standardized as much as possible, it should be noted that it depends on the inspector's subjective assessment. What one inspector considers a defect, another may consider acceptable. Therefore, different inspectors may evaluate the same roll of fabric very differently, regardless of its actual quality.

[0005] The use of technology has improved fault detection methods at various stages of fabric manufacturing. Efficient image capture and image analysis techniques enable the inspection of woven fabrics.

[0006] For example, the international patent publication number WO2006117673, granted to Gironi Pietro, entitled "Apparatus and method for in-line reading and control of warp threads in a loom," describes an apparatus and method for reading and controlling warp threads that uses a device to read an image and compare the acquired image with one or more predetermined samples to determine defects in the work cycle, thereby immediately interrupting the operation of the loom in response to the determined defects.

[0007] In another example, U.S. Patent No. 9,909,238, granted to Wolf Markus and Ackermann Armin, entitled "Monitoring device for a weaving machine, weaving machine, and method for monitoring," describes a monitoring device including a camera and a weft-beating device. The weft-beating device includes a reed or slat extending along the weft-beating device. The camera is fixed to the weft-beating device and includes adjacent sensor elements arranged in a row extending parallel to the longitudinal direction of the weft-beating device.

[0008] In yet another example, U.S. Patent No. 5,165,454, granted to Kabushiki Kaisha Toyoda Jidoshokki Seisakusho and Kabushiki Kaisha Toyota Chuo Kenkyusho, entitled "Detection of warp in reed dent before loom start-up," describes a warp insertion monitoring method and apparatus for protecting woven fabric from defects caused by incorrect or failed warp insertion. A warp detector on the loom detects the presence or absence of warp yarns, thereby identifying anomalies in the position where the warp yarns pass through the reed. Specifically, in this system, the timing of warp detection is specifically chosen during a period when the loom is stopped, so that the presence or absence of errors can be detected before the loom is restarted.

[0009] An improved technique is still needed to detect faults quickly and cost-effectively using fabric inspection systems on looms. The system and method described in this paper aim to address this need. Summary of the Invention

[0010] One aspect of the invention is to provide an inspection system for a loom, the system comprising: at least one imaging device configured to collect images of at least a portion of the weaving area of ​​the loom; at least one image processor configured and operable to detect irregularities in the image data; and at least one frame grabber configured and operable to receive images of at least one feed-pick from the imaging device and to send compressed image data packets to the image processor; wherein the compressed image data packets contain a characteristic sequence of warp floats and warp sinks along the feed-pick. Optionally, the system may further include at least one image capture trigger mechanism operable to trigger the imaging device to capture desired instantaneous images during a weaving cycle.

[0011] In another aspect of the invention, a method for inspecting woven fabrics is taught. The method may include providing a fabric inspection system on a loom; obtaining a reference matrix representing a desired woven pattern, the reference matrix comprising a two-dimensional array of values ​​arranged as a row sequence, each row corresponding to a series of desired warp floats and desired warp sinks along a single weft yarn; capturing an image of the weft yarn along the weft line of the woven area; identifying a characteristic sequence of warp floats and warp sinks along the weft yarn in the image; generating a string of numbers corresponding to the characteristic sequence; and comparing the string of numbers with the corresponding row of the reference matrix.

[0012] Typically, a string of numbers contains a sequence of binary or Boolean values. Additionally or alternatively, the string of numbers may include a sequence of values ​​that further indicate the color.

[0013] Where appropriate, the step of capturing the weft image further includes: capturing an image of at least a portion of the woven area; transmitting the image data to at least one image processor; and identifying the weft yarn in the image data. Optionally, the imaged portion of the woven area includes all shed areas, woven fabric areas, and weft edges.

[0014] Additionally or alternatively, the method further includes: providing at least one imaging device configured to collect images of at least a portion of the weaving area of ​​the loom; providing at least one image capture triggering mechanism; selecting a desired moment during the weaving cycle; and the at least one image capture triggering mechanism triggering the imaging device at the desired moment of the weaving cycle. Thus, the desired moment can be selected to coincide with the moment the shed opens.

[0015] Optionally, the method further includes generating a precision metric based on the deviation of the numerical string from the corresponding row of the reference matrix. Optionally, the precision metric indicates the presence of weaving defects. Alternatively, or additionally, the method may include generating a standard quality index for the woven fabric.

[0016] When the accuracy metric exceeds a threshold, the method can further initiate an automatic correction process if necessary. For example, the automatic correction process can be selected from at least one of the following groups: stopping the loom, releasing the fabric, adjusting the weft insertion force, generating an alarm, etc., and combinations thereof.

[0017] In a different manner, the step of obtaining the reference matrix includes accessing a reference pattern stored in a memory component. Alternatively, the step of obtaining the reference matrix may include: monitoring the ongoing weaving process; identifying the repetition cycle in the weaving process; generating a reference matrix based on the repetition cycle; and storing the reference matrix in a memory component.

[0018] A particular aspect of this disclosure is to teach a method that further includes: providing at least one imaging device configured to collect images of at least a portion of a weaving area of ​​a loom; providing a frame capture device configured and operable to receive images from the imaging device; providing an image processor; and sending compressed image data packets to the image processor.

[0019] Typically, compressed image data packets contain a sequence of Boolean values ​​representing a characteristic sequence of warp floats and sinks along the weft yarn. Additionally or alternatively, compressed image data packets contain a sequence of values ​​representing a portion of the captured image, which includes only a reduced portion of the shed area, the weft yarn, and a portion of the shed area. Attached Figure Description

[0020] To better understand the embodiments and show how the invention can be implemented, reference will now be made to the accompanying drawings by way of example only.

[0021] Referring now in detail and specifically to the accompanying drawings, it should be emphasized that the details shown are merely exemplary and are intended only for illustrative discussion of selected embodiments, and are given to provide a description of principles and concepts that are believed to be most useful and readily understood. In this regard, no attempt is made to provide more detailed structural details than necessary for a basic understanding; the description taken in conjunction with the drawings makes it apparent to those skilled in the art how to put the various selected embodiments into practice.

[0022] In the attached diagram:

[0023] Figure 1 A schematic side view of an exemplary configuration of a fabric inspection system integrated into a loom is shown;

[0024] Figure 2 It shows Figure 1 A schematic side view of a fabric inspection system, in which a focusing image capture device is used to capture images of the weave area and the newly woven fabric;

[0025] Figure 3It shows Figure 1 A schematic side view of a fabric inspection system, in which a focusing image capture device is used to capture images of the warp yarns in the shed;

[0026] Figure 4 This is a block diagram showing the main components of a first embodiment of a fabric inspection system on a loom;

[0027] Figure 5 A schematic side view of an exemplary configuration of the fabric inspection system on a loom according to the present invention is shown;

[0028] Figure 6A and Figure 6B It shows Figure 5 A schematic side view of a fabric inspection system, in which a focusing image capture device is used to capture images of the woven area;

[0029] Figure 7 This is a flowchart illustrating a method for detecting defects in woven fabrics using a fabric inspection system on a loom;

[0030] Figure 8A It is a representation of a frame imaged by the image capture device of the fabric inspection system on the loom;

[0031] Figure 8B A cross-sectional view showing the weft yarns interlaced between a set of warp threads;

[0032] Figure 8C This represents a one-dimensional Boolean array, which represents the characteristic sequence after floating-point and sinking-point operations;

[0033] Figure 9A Indicates a woven pattern;

[0034] Figure 9B Indicates corresponding to Figure 9A Reference matrix of the woven pattern;

[0035] Figure 9C Indicates following Figure 9A Woven fabrics with woven patterns;

[0036] Figure 10A and Figure 10B This indicates the weft sequence added next to their corresponding rows in the reference matrix; and

[0037] Figure 11 This is a flowchart of a method for detecting anomalies during weaving. Detailed Implementation

[0038] Various aspects of this disclosure relate to systems and methods for inspecting fabrics on looms.

[0039] Detailed embodiments of the invention are disclosed herein as needed; however, it should be understood that the disclosed embodiments are merely examples of the invention, which may be implemented in various and alternative forms. The drawings are not necessarily drawn to scale; certain features may be enlarged or minimized to show details of specific components. Therefore, the specific structural and functional details disclosed herein should not be construed as limiting, but merely as a representative basis for teaching those skilled in the art to use the invention in various ways.

[0040] Suitablely, in various embodiments of this disclosure, one or more tasks described herein may be performed by a data processor, such as a computing platform or distributed computing system for executing multiple instructions. Optionally, the data processor includes or accesses volatile memory for storing instructions, data, etc. Additionally or alternatively, the data processor may access non-volatile memory for storing instructions and / or data, such as magnetic hard disks, flash drives, removable media, etc.

[0041] It should be noted that the systems and methods disclosed herein may not be limited to the details of the construction and arrangement of the components or methods set forth in the specification or shown in the drawings and examples. The systems and methods of the present invention can have other embodiments, or can be practiced and implemented in various ways and techniques.

[0042] Alternative methods and materials similar to or equivalent to those described herein may be used to practice or test embodiments of this disclosure. However, the specific methods and materials described herein are for illustrative purposes only. The materials, methods, and embodiments are not necessarily limiting. Therefore, various processes or components may be appropriately omitted, substituted, or added in various embodiments. For example, method steps may be performed in a different order than described, and various steps may be added, omitted, or combined. Furthermore, aspects and components described with respect to certain embodiments may be combined in various other embodiments.

[0043] Figure 1 This illustrates an exemplary configuration of a fabric inspection system 100 on a loom. The loom 102 includes a yarn roller 106, a take-up roller 120, a pair of heald frames 108A and 108B, and a reed 110. An array of warp yarns 104 passes through the heald frames 108A and 108B and the reed 110. The heald frames 108A and 108B are made of wood or metal such as aluminum. They carry a number of heddles (not shown), through which the ends of the warp yarns pass. The heald frames 108A and 108B are configured to raise and lower the warp yarns, thereby creating a shed 112 through which weft yarns (not shown) can be inserted using some weft insertion mechanism (not shown) such as a shuttle, rapier, nozzle, etc. The reed 110 is a metal comb used to press the weft yarns against the newly woven fabric 116. It also helps to maintain the position of the warp yarns 104. The woven fabric 116 is collected by the take-up roller 120 during its production.

[0044] A fabric inspection system 100 on a loom is configured to monitor a knitted area 118, which includes newly knitted fabric 116, a shed 112, and a weft area 114. The weft area 114 is a segment of the knitted area 118 where, during operation of the loom 102, the reed 110 strikes the weft yarn along the weft line. The weft line is the boundary beyond which the already knitted fabric 116 extends. The fabric inspection system 100 includes one or more image capture devices 122 in communication with an image processor 124. Exemplary image capture devices 122 include analog or digital still image cameras, video cameras, optical cameras, laser cameras, laser or 3D image scanners, or any other device capable of capturing high-resolution images of the knitted area 118. The image capture device 122 may also be a communication device such as a computer, laptop computer, or mobile phone with a high-definition built-in camera. In an exemplary embodiment, to capture images of a loom operating at high speed, the required camera needs to be very high-speed, for example, capturing more than 1000 frames per second. The image processor 124 is operable to receive and process data collected by the image capture device 122. The image processor 124 can be a server computer, client user computer, personal computer (PC), tablet PC, laptop computer, desktop computer, mobile phone, control system, and network router, switch, or bridge. Alternatively, the image processor 124 can be a software application running in a virtual cloud environment. An output mechanism 126 associated with the image processor 124, such as a visual display unit, can provide the user with information about the functionality of the loom 102 and when any malfunction is detected. The information can be provided in the form of images, graphical representations, numbers, or text, and can be associated with measurement data, statistics, etc. The output mechanism 126 can also display alarms or indicators in the event of any deviation from the normal operation of the loom 102. It should be noted that this configuration of the fabric inspection system 100 on the loom is operable to monitor the knitting area 118 during the operation of the loom 102. Therefore, a computer can be connected to the loom 102 and operable to stop the loom 102 or otherwise adjust the settings of the loom 102 in response to data collected from the monitored knitting area 118.

[0045] Typically, when heald frames 108A and 108B are separated, the fabric inspection system 100 on the loom captures an image of the weaving area 118. In this state, since the warp yarns 112 and the weft area 114 are not coplanar, it is impossible to simultaneously focus on the warp yarns 112 and the weft area 114 in the shed. Therefore, it is necessary to adjust the object distance of the image capturing device 122 to capture an image of either the warp yarns 112 or the weft area 114. Figure 2An exemplary embodiment is shown, wherein the image capturing device 212 focuses to capture images of the weave region 206 and region 208 of the newly woven fabric. In this case, the image capturing device 212 cannot capture images of the warp yarns 202A and 202B in the shed region 204. Figure 3 Another exemplary embodiment is shown, wherein the image capturing device 312 is focused to capture images of the warp yarns 302A, 302B in the shed region 304. In this case, the image capturing device 312 cannot capture images of the weave region 306 and region 308 of the newly woven fabric.

[0046] As a remedy, multiple image capture devices can be used, focusing on different areas 204, 206, 208 (or 304, 306, 308); however, this increases the cost and time of analyzing individual images.

[0047] Now for reference Figure 4 The diagram illustrates the main components of a fabric inspection system 400 on a loom according to the present invention. System 400 can identify faults during the fabric manufacturing process, thereby enabling early detection or prevention of fabric defects. For example, the loom system 400 described herein can be used as a cost-effective tool for providing continuous monitoring of woven fabrics during production and can provide industry standards for quality control of such fabrics.

[0048] The fabric inspection system 400 on a loom includes an image capture triggering mechanism 406, an image capture device 408, an image processor 410, a controller 412, and an output mechanism 414. The image capture triggering mechanism 406 is configured to trigger the image capture device 408 based on desired conditions. The image capture device 408 is configured to collect image data from the weaving area 402 of the loom 404 and transmit the data to the image processor 410.

[0049] Various types of image capture devices 408 suitable for this requirement can be used. Exemplary image capture devices 408 include analog or digital still image cameras, video cameras, optical cameras, laser cameras, laser or 3D image scanners, or any other device capable of capturing high-resolution images of the woven area 402. Image capture device 408 can also be a communication device such as a computer, laptop computer, or mobile phone with a high-definition built-in camera. In an exemplary embodiment, in order to capture images of a loom operating at high speed, the required camera needs to be very high-speed, for example, capturing more than 1000 frames per second. Furthermore, array cameras, etc., with a resolution suitable for detecting individual yarns within the woven fabric can be used. The resolution of the image capture device 408 can be selected based on the cost and properties of the fabric being inspected. The resolution can be less than 1 mm, for example, approximately 0.1 mm as required.

[0050] Image capture triggering mechanism 406 may include a detector or sensor connected to loom 404 and configured to detect heald frames 508A and 508B of loom 404 (e.g., Figure 5 The image capture device 408 can be triggered by a detector when the heald frame meets the required conditions. Exemplary detectors may include mechanical, electrical, or optical sensors. It should be noted that the scope of the invention is not limited to the exemplary detectors described above, and any other detector capable of detecting heald frame movement may be used for this purpose.

[0051] In another embodiment, the image capture triggering mechanism 406 may additionally or alternatively include a timer, such as a strobe or a lamp, which can be timed to produce a flash when the frame meets the desired conditions.

[0052] In other embodiments, the image capture trigger mechanism 406 may additionally or alternatively include a receiver that communicates with the loom 404 and is configured to receive output signals from the encoder of the loom motor. For example, a communication cable may be connected between the output terminals of the loom 404 and the input terminals of the image capture trigger mechanism 406. Thus, when desired conditions are met, a trigger signal can be sent; for example, the image capture trigger mechanism 406 may receive a shuttle-throwing signal indicating that the shuttle-throwing process has been initiated, and the shuttle-throwing signal can be used as a trigger signal for the image capture device 408.

[0053] Image data collected by image capture device 408 is sent to image processor 410, which analyzes the received image data and identifies irregularities indicating weaving faults. System 400 can use a variety of image processors 410. Processors such as computers, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), and microprocessors can be selected to provide image processing at a sufficiently fast rate. The processing rate can be fast enough to allow real-time analysis of each frame imaged by image capture device 408. Optionally, image processor 410 is operable to segment each frame and analyze each frame segment individually and possibly at a separate sampling rate. Exemplary image processors 410 include server computers, client user computers, personal computers (PCs), tablet PCs, laptop computers, desktop computers, mobile phones, control systems, and network routers, switches, or bridges. Alternatively, image processor 410 can be a software application running in a virtual cloud environment.

[0054] A controller 412 is provided to respond to the detection of a knitting fault. The controller 412 may respond, for example, by outputting data to an output mechanism 414. Associated with the image processor 124, the output mechanism 414, such as a vision display unit, can provide the user with information about the functionality of the loom 102 and when any fault is detected. The information may be provided in the form of images, graphical representations, numbers, or text, and may be associated with measurement data, statistics, etc. The output mechanism 414 may also display an alarm or indicator if any deviation from the normal operation of the loom 404 is detected. The output mechanism 414 may also include a database to store processed image data. If necessary, the controller 412 may further operate to activate an overrun switch 416 in response to the detection of a defect to stop or otherwise adjust the loom 404. The overrun switch 416 may be an actuator or any other system suitable for this requirement.

[0055] In one embodiment of the present invention, the image capture triggering mechanism 406 is adjusted such that when frames 508A and 508B (e.g.) Figure 5 When the warp yarns (as shown) are aligned with each other, the image capture device 408 is triggered, and in this case, the image capture device 408 is triggered to capture an image of the weaving area 402. In this case, the warp yarns in the shed are coplanar with the weaving area and the new weave.

[0056] Now for reference Figure 5 It shows a schematic side view of an exemplary configuration of a fabric inspection system 500 according to the invention, integrated onto a loom 502.

[0057] Figure 5 The construction of the 502 loom is similar to Figure 1 Except that heald frames 508A and 508B are depicted at the same height and aligned with each other, the upper and lower warp yarns in shed 512 are in the same plane as the weave region 514 of the fabric and newly woven fabric 516. Image capture of the weave region 518 allows a single object distance of image capture device 522 to be used to image both the shed region 512 and the weave region 514, thereby allowing the detection of irregularities in both regions. Detector 528 is included in system 500 for this purpose. Preferably, images of the weave region 518 are captured twice in each movement cycle (up and down) of heald frames 508A and 508B to capture two sets of warp yarns. Detector 528 may include sensors such as mechanical sensors, electrical sensors, optical sensors, etc., and combinations thereof.

[0058] Figure 6AA schematic side view of a fabric inspection system 600 is shown, in which a focused image capture device 612 is used to capture an image of the woven area 614. Since the upper and lower warp yarns in the shed 604 are in the same plane as the weave area 606 and the new woven fabric 608, the image capture device 612 can use a single depth of focus on a wide-angle 610 to capture an image of the entire woven area 614.

[0059] Figure 6B Another configuration of the fabric inspection system 600 is shown, in which the heald frame is separated to raise the upper warp 602A and lower the lower warp 604B, thereby forming a shed. It is particularly noteworthy that, where appropriate, images can be captured additionally or alternatively in this configuration. Thus, the image capture device 612 can image only the upper warp 602A, thereby allowing the image processor 410 ( Figure 4 (As shown in the diagram) it makes it easier to distinguish the warp floats and warp sinks along the weft yarn.

[0060] Return to reference Figure 5 In yet another alternative embodiment, the image capture triggering mechanism 529 may be selected to trigger the image capture device 522 in other ways. For example, the image capture triggering mechanism 529 may include a timer 527, allowing a fixed time to be set for the shutter of the image capture device 522 to capture an image of the woven area 518. The shutter timing may be set to an instance where the heald frames 508A and 508B are aligned with each other. In this case, the image of the woven area 518 is captured without being triggered by the detector 528.

[0061] Additionally or alternatively, the image capture trigger mechanism 529 may also include a receiver 523 that communicates with the loom 502 and is configured to receive an output signal from the encoder of the loom motor 503.

[0062] Now for reference Figure 7 The flowchart illustrates exemplary method steps of an embodiment of the present invention for detecting defects in woven fabrics using a fabric inspection system 500 on a loom.

[0063] In step 702, a fabric inspection system 500 is provided on the loom. Optionally, during operation of the loom 502, in step 704, an image capture triggering mechanism 529, which may include a detector 528, may monitor the positions of heald frames 508A and 508B. At step 706, for example, when heald frames 508A and 508B are aligned with each other, an image capture device 522 is triggered at a desired point in the cycle. Then, in step 708, the image capture device 522 collects images of the weave area 518, including the shed 512, the weave area 514, and the newly woven fabric 516.

[0064] In step 710, the image data is transmitted to the image processor 524. In step 712, the image processor 524 analyzes the irregularities and faults in the received image data. In step 714, if the irregularities detected in the image data indicate that a weaving fault has occurred, the fault is recorded on the output mechanism 526 in step 716. This process can be continued by collecting and analyzing another image, thus allowing the process to be repeated.

[0065] Optionally, the fabric inspection system 500 on the loom may further include a frame grabber 532, which is configured and operable to receive images from the imaging device 522 and send compressed image data packets 534 to the image processor 524.

[0066] It should be noted that the record of faults can include simple fault counts, such as using a deduction system like four points. Alternatively, more precise data, such as those relating to the types of faults detected and their statistical distribution, can be recorded.

[0067] refer to Figure 8A The image frame 800 shows a representation of a woven area 808 imaged by the image capture device 522 of the fabric inspection system 500 on the loom. Frame 800 shows the shed 802, the weave area 804, and the new woven fabric 806. It also shows oil stains caused by contamination portions 810 spreading along the new woven fabric 806. Image frame 800 is processed by image processor 524 to detect contamination portions 810, and the loom operator can take appropriate measures to resolve the problem.

[0068] Weaving defects can occur in any of these areas of frame 800 and can be detected using the fabric inspection system 500 on the loom. For example, roving knots, missed yarns, and missing warp threads can be detected in the shed 802 and weft area 804, while oil stains, loom stop marks, and start-up marks can be detected in the new woven fabric 806.

[0069] Various defects that may occur in the weaving area 808 during manufacturing can lead to defects in the finished fabric. These include roving knots, holes, loose yarns, yarn variations, missing warp, stained yarns, incorrect yarn defects, oil stains, loom stop marks, start marks, thin sections, frayed edges, reed marks, mixed weft yarns, twisted weft yarn defects, mixed warp yarns, long knots, pulled-in defects, low weft, broken weft yarns, double weft, double warp, elastic band defects, spot defects, etc. It should be noted that the listed defects are exemplary in nature and should not be used to limit the scope of the invention.

[0070] In other embodiments of fabric inspection, a novel method for identifying defects row by row can be used. According to this method, a reference pattern representing the desired pattern of the fabric can be obtained. Such a reference pattern can, for example, be converted into a two-dimensional matrix comprising an array of values ​​distributed in rows and columns.

[0071] For example, in a reference pattern used for weaving, each column of the array can correspond to the warp end, and each row can correspond to the weft or a single weft that is inserted through the shed during the shuttle insertion to intersect with the warp.

[0072] Although only the weaving pattern is described here for the sake of brevity, it should be noted that this inspection system can be used with other types of fabrics, such as tufted fabrics.

[0073] It is important to note that the reference matrix can be composed of Boolean values ​​when needed. Therefore, for example, in a woven pattern, a 0 (ZERO) value can be used to indicate a warp float, where the warp yarns are above the weft yarns, while a 1 (ONE) value can be used to indicate a warp sink, where the warp yarns are below the weft yarns. Alternatively, a 0 (ZERO) value can be used to indicate a warp sink, and a 1 (ONE) value can be used to indicate a warp float.

[0074] Refer again Figure 8A Using a fabric inspection system such as that described herein, images of the shed 802, the weft insertion area 804, and the new knitted fabric 806 are collected. Therefore, the weft insertion weft yarn 805 can be identified before or after weft insertion during each cycle of the loom. As used herein, the term weft insertion weft yarn refers to the newest weft yarn to be inserted into the shed, and is therefore the weft yarn furthest from the knitted fabric.

[0075] refer to Figure 8B This schematically represents a portion of the weft yarn 805 interwoven between a set of warp threads 812. A particular feature of the current method is the ability to analyze an image of the weft yarn 805 to identify the characteristic sequence of its warp floats 801 and warp sinks 803. For example... Figure 8C As shown, this characteristic sequence via floating point 801 and sinking point 803 can be represented by a one-dimensional Boolean array or a number string 820.

[0076] It is important to note that when the shed is in an open configuration, the contrast between the warp float and warp sink can be enhanced; therefore, it may be helpful to schedule image collection to coincide with the points in the weaving cycle when the shed is open. It should also be noted that the contrast between the warp and weft threads can be further enhanced by adjusting the illumination between the upper and lower shed lights as needed.

[0077] Alternatively or additionally, as described herein, it may be necessary to capture images of the weft yarn at points in the cycle where the warp yarns in the shed are coplanar with the weft area and the new woven fabric.

[0078] The feature sequence of the imaged weft yarn can be compared with the corresponding row of a reference matrix to generate a precision metric. This precision metric can be used to indicate the presence of weaving defects and can be used in a defect calculation function to generate a standard quality index for the woven fabric.

[0079] When the accuracy measurement exceeds the threshold, the automatic process can be started in a non-restrictive manner, such as stopping the loom, taking the fabric off the loom, adjusting the force for the next weft insertion cycle, generating an alarm, etc.

[0080] For example, a measure of precision can be determined by counting the number of errors that occur when the floating-point or sinking-point values ​​do not match the corresponding values ​​in a reference matrix. Error density can be determined, for example, by counting the number of errors within a given length of fabric or within a given number of weft yarns. Therefore, fabrics can be graded based on error density, where fabric with fewer than one error in 100,000 weft yarns is of higher quality than fabric with fewer than one error in 50,000 weft yarns.

[0081] Additionally or alternatively, a precision metric can be a weighted score, assigning larger values ​​to errors that are close to each other, rather than those that are more widely spaced. For example, a precision metric can be calculated as follows:

[0082]

[0083] Where AM represents the precision metric, and E j W represents the weighted error when the value transmitted via floating-point or sinking-point methods does not match the corresponding value in the reference matrix. j E represents j The weighting coefficients can vary depending on how close the detected error is to the actual error.

[0084] The value of the precision metric itself can be used as an input parameter for a defect calculation function, which can be used in conjunction with other quality indicators, such as weft pitch function, low weft count, missing yarn count, roving knot count, oil stain count, loom stop count, as described in, for example, U.S. Patent No. 9,499,926, the entire contents of which are incorporated herein by reference, or other defects that may be conceived by those skilled in the art.

[0085] For example, the quality index can be determined by summing the terms representing the accuracy of the weave pattern, fault detection, and weft spacing using the following quality function:

[0086]

[0087] Where Q represents the calculated quality index value, K E The weighting factor, F, represents the accuracy metric. i W represents the count of a specific fault. iRepresents a specific fault type F i The weighting coefficients, N f K represents the number of recorded fault counts. f S represents the weighting factor that accounts for the contribution of the fault count to the quality index. k The value representing the weft yarn spacing, W k S represents the spacing value of each weft yarn. k The weighting coefficients, and K f The weighting factor representing the contribution of weft pitch deviation to the quality index.

[0088] It should be noted that the above examples of calculations of quality functions and accuracy measures are provided for illustrative purposes only, and that other quality functions that may be conceived by those skilled in the art may be used additionally or alternatively.

[0089] Now for reference Figure 9A An exemplary desired weave pattern is shown, which includes desired warp float 971 and desired warp sink 973. For example, a basket weave is shown for illustrative purposes. The weave pattern can be converted into, for example... Figure 9B The reference matrix 900 is shown. Reference matrix 900 is a two-dimensional array of Boolean values, where each row 901-912 corresponds to the weft yarn. This pattern will produce... Figure 9C The woven fabric exhibits the characteristics shown.

[0090] Now for reference Figure 10A and Figure 10B The image shows a series of weft yarns 951-962 added to the woven fabric. As each weft yarn is added, it can be imaged, allowing the image to be compared with data from... Figure 9B The corresponding row of the reference matrix 900 is compared. Therefore, when the first weft yarn 951 is added, it can be imaged and compared with the string 901 of the first row of the corresponding first row 901 of the reference matrix 900.

[0091] The comparison is performed using {1,1,1,0,0,1,0,1,1,1,0}. Similarly, the second weft yarn 952 is compared with string 902, and so on, until the entire fabric is produced. In this way, the fabric can be inspected during production and graded on the loom.

[0092] Now for reference Figure 11 The flowchart illustrates a possible method for detecting defects using one-dimensional inspection analysis, as described herein. This method includes the following steps:

[0093] A reference pattern 1102 is obtained, for example, by referring to a pattern stored in a memory component. Alternatively or additionally, the reference pattern can be generated by learning on a loom in repeated cycles, and then stored in memory for the processor to reference.

[0094] The reference pattern is converted into a reference matrix 1104, which typically consists of an array of Boolean values; however, when examining more complex patterns, such as those using colors, other arrays may be preferred.

[0095] Image 1106 of the weave line is obtained, preferably from a photograph of all three, including the shed area, the weave area, and a portion of the woven fabric.

[0096] The image of the weft thread is used to identify where there are warp floats and warp sinks along the weft thread, thus generating a feature sequence 1108. A numerical string corresponding to the feature sequence is generated 1110. The numerical string used for the feature sequence is compared with the corresponding row of the reference matrix 1112. If a difference is detected 1114, the defect can be appropriately recorded 1116.

[0097] The technical and scientific terms used herein should have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure pertains. However, it is anticipated that many related systems and methods will be developed during the patent life of this application. Therefore, the scope of terms such as computing unit, network, display, memory, server, etc., is intended to include all these previously known new technologies.

[0098] The terms “including,” “comprising,” “covering,” “inclusive,” “having,” and their combinations as used herein mean “including, but not limited to,” and indicate that the listed components are included, but do not exclude other components in general. Such terminology encompasses the terms “including” and “primarily including.”

[0099] As used in this specification, the singular indefinite articles “a,” “an,” and the definite article “the” should be considered to include or additionally cover single and multiple indicators, unless the content clearly indicates otherwise. In other words, these terms apply to one or more objects. For example, the terms “compound” or “at least one compound” can include multiple compounds, including mixtures thereof. As used in this specification, the term “or” is generally used to include or additionally cover “and / or”, unless the content clearly indicates otherwise.

[0100] The word “exemplary” is used herein to mean “serving as an example, instance, or illustration.” Any embodiment described as “exemplary” is not necessarily to be construed as preferred or advantageous over other embodiments, or to exclude the inclusion of features in other embodiments.

[0101] The term "optionally" is used herein to mean "provided in some embodiments but not in others." Any specific embodiment of the invention may include a number of "optional" features unless these features conflict.

[0102] It should be understood that, for clarity, certain features of the invention described in the context of a single embodiment may also be provided in combination in that single embodiment. Conversely, for clarity, various features of the invention described in the context of a single embodiment may also be provided individually or in any suitable sub-combination or as suitably provided in any other described embodiments of the invention. Certain features described in the context of various embodiments are not considered essential features of those embodiments unless the embodiment does not function without these elements.

[0103] Although the invention has been described in conjunction with specific embodiments therein, other substitutions, modifications, variations, and equivalents will obviously be apparent to those skilled in the art. Therefore, it is intended to cover all such substitutions, modifications, variations, and equivalents falling within the spirit of the invention and the broad scope of the appended claims. Furthermore, the various embodiments set forth above are described with reference to exemplary block diagrams, flowcharts, and other illustrations. It will be apparent to those skilled in the art that the illustrated embodiments and their various alternatives can be implemented without being limited to the illustrated examples. For example, the block diagrams and the appended description should not be construed as requiring a particular architecture, layout, or configuration.

[0104] The presence of broadening words and phrases such as “one or more,” “at least,” “but not limited to,” or other similar phrases should not, in some cases, be construed as implying a narrower scope expected or required where such broadening phrases may not exist.

[0105] Furthermore, embodiments may be implemented using hardware, software, firmware, middleware, microcode, hardware description languages, or any combination thereof. When implemented in software, firmware, middleware, or microcode, program code or code segments that perform the necessary tasks may be stored in a computer-readable medium such as a storage medium. The processor can then perform the necessary tasks.

[0106] All publications, patents, and patent applications mentioned in this specification are incorporated herein by reference in their entirety, to the same extent that each individual publication, patent, or patent application is specifically and individually incorporated herein by reference. Furthermore, any reference or designation of any reference in this application should not be construed as an admission that such references are available as prior art to the invention. The use of section headings should not be construed as a necessary limitation. The scope of the disclosed subject matter is defined by the appended claims and includes combinations and sub-combinations of the various features described above, as well as variations and modifications that can be obtained by one skilled in the art upon reading the foregoing description.

Claims

1. A method for inspecting a knitted fabric comprising: providing a fabric inspection system (500) on a loom, obtaining a reference matrix (900) representing a desired knitting pattern, the reference matrix (900) comprising a two-dimensional array of values arranged in a sequence of rows, each row corresponding to a series of desired warp floats (801) and desired warp sinkers (803) along a single weft yarn; capturing an image of a shed weft yarn (805) along a shed line of a knitting area (518); identifying a sequence of warp float (801) and warp sinker (803) features along the shed weft yarn (805) in the image; generating a digital string (820) corresponding to the sequence of features; and comparing the digital string (820) to a corresponding row (901) of the reference matrix (900); and generating an accuracy measure based on a deviation of the digital string (820) from the corresponding row (901) of the reference matrix (900).

2. The method of claim 1, wherein the step of capturing the image of the shed weft yarn (805) further comprises: capturing an image of at least a portion of a knitting area (518); transferring image data to at least one image processor (524); and identifying the shed weft yarn (805) within the image data.

3. The method of claim 2, wherein the at least a portion of a knitting area (518) comprises all of an interlacing area (512), a knitted fabric area, and a shed area (514).

4. The method of claim 1, wherein the digital string (820) comprises a sequence of binary or Boolean values.

5. The method of claim 1, wherein the digital string (820) comprises a sequence of values further indicating color.

6. The method of claim 1, further comprising providing at least one imaging device (522) configured to collect an image of at least a portion of a knitting area (518) of a loom; providing at least one image capture trigger mechanism (529); selecting a desired instant during a knitting cycle; and the at least one image capture trigger mechanism (529) triggering the imaging device (522) at the desired instant during a knitting cycle.

7. The method of claim 6, wherein the desired instant coincides with a moment when an interlacing (112) is open.

8. The method of claim 1, wherein the accuracy measure indicates a presence of a knitting defect.

9. The method of claim 1, further comprising generating a standard quality index for the knitted fabric.

10. The method of claim 1, further comprising initiating an automatic correction process when the accuracy measure exceeds a threshold value.

11. The method of claim 10, wherein the automatic correction process is selected from at least one of the group consisting of: stopping the loom, removing the cloth from the loom, adjusting a beat-up force, generating an alarm, and combinations thereof.

12. The method of claim 1, wherein the step of obtaining a reference matrix (900) comprises accessing a reference pattern stored in a memory component. ​ 13. The method of claim 1, wherein the step of obtaining a reference matrix (900) comprises: monitoring an ongoing weaving process; identifying a repeating cycle in the weaving process; generating the reference matrix (900) from the repeating cycle; and storing the reference matrix (900) in a memory component.

14. The method of claim 1, further comprising: providing at least one imaging device (522) configured to collect images of at least a portion of a weaving area (518) of a loom; and providing a frame grabber (532) configured and operable to receive images from the imaging device (522); providing an image processor (524); and sending compressed image data packets to the image processor (524).

15. The method of claim 14, wherein the compressed image data packets comprise a sequence of Boolean values representing the sequence of features of a float (801) and a sink (803) along the shed weft (805).

16. The method of claim 14, wherein the compressed image data packets contain a sequence of values representative of a portion of a captured image that includes only a reduced portion of a shed area, a shed weft (805), and a portion of a shed area.

17. The method of claim 3, further illuminating the shed area (512) from below.

18. The method of claim 3, further illuminating the shed area (512) from above.

19. A fabric inspection system (500) on a loom, comprising: at least one imaging device (522) configured to collect images of at least a portion of a weaving area (518) of a loom (502); at least one image processor (524) configured and operable to detect irregularities in image data; at least one frame grabber (532) configured and operable to receive images of at least one shed weft (805) from the imaging device (522) and send compressed image data packets to the image processor (524); wherein the compressed image data packets comprise a sequence of features of a float (801) and a sink (803) along the shed weft (805), and a memory for storing a reference matrix (900) representing a desired weaving pattern, the reference matrix (900) comprising a two-dimensional array of values arranged as a sequence of rows, each row corresponding to a sequence of desired floats (801) and desired sinks (803) along a single weft; wherein the at least one image processor (524) is configured and operable to: generate a numeric string (820) corresponding to the sequence of features; compare the numeric string (820) to a corresponding row (901) of the reference matrix (900); and generate an accuracy measure based on a deviation of the numeric string (820) from the corresponding row (901) of the reference matrix (900). ​ ​ 20. The on-loom inspection system (500) of claim 19, further comprising at least one image capture trigger mechanism (529) operable to trigger the imaging device (522) to capture an image at a desired instant during a weaving cycle.

Images