A visual sensor-based detection method for broken needles in a flat knitting machine

Abstract

The present application belongs to the field of flat knitting machines, and particularly relates to a flat knitting machine needle breakage detection method based on a visual sensor, which comprises the following steps: S1: a needle breakage image template of a flat knitting machine needle is pre-photographed and uploaded to a computer; S2: a visual sensor is used to photograph a needle tip of the flat knitting machine needle, obtain needle tip image data, and transmit the needle tip image data to the computer; S3: the computer compares the needle tip image data with the needle breakage image template, and judges whether needle breakage occurs; and S4: when the computer identifies and judges that needle breakage occurs, the computer outputs prompt information to a worker. The above setting is used to complete automatic detection of needle breakage of the flat knitting machine needle, so that the worker can be informed of the occurrence of needle breakage in a timely manner, and the worker can continue and handle in a timely manner.

Description

Technical Field

[0001] This invention relates to the field of flat knitting machines, specifically a method for detecting broken needles in flat knitting machines based on a vision sensor. Background Technology

[0002] A flat knitting machine, a type of warp knitting machine, is primarily used for manufacturing knitted fabrics. It differs from traditional warp knitting machines in its knitting method and application. Flat knitting machines primarily use hooks to pull yarn from one needle cylinder to another to form fabric. This knitting method gives fabrics produced by flat knitting machines high elasticity, thickness, softness, and good drape. Flat knitting machines are suitable for knitting blended yarns such as wool, rabbit hair, camel hair, cashmere, silk, and synthetic fibers. They can knit various knitted products including sweaters, cashmere sweaters, wool pants, scarves, and hats in single and double-sided plain weave, rib, two-tone, ribbed, twisted rope, and pique patterns. Flat knitting machines are widely used in homes, yarn and sweater shops, sweater factories, scarf and hat manufacturers, and clothing and apparel businesses.

[0003] However, during the flat knitting process, yarn breakage can occur, requiring timely reconnection by staff. Otherwise, the quality of the textiles and the knitting process will be affected. However, existing staff often find it difficult to detect yarn breakage in a timely manner. Therefore, a method for detecting broken needles on flat knitting machines based on visual sensors is proposed to address this problem. Summary of the Invention

[0004] In order to overcome the shortcomings of the prior art and solve at least one of the technical problems mentioned in the background art, the present invention proposes a method for detecting broken needles on a flat knitting machine based on a vision sensor.

[0005] The technical solution adopted by this invention to solve its technical problem is: a broken needle detection method for a flat knitting machine based on a vision sensor, which includes the following steps:

[0006] S1: Take a pre-shot image template of a broken needle on a flat knitting machine and upload it to the computer;

[0007] S2: Use a vision sensor to photograph the needle tip of the flat knitting machine, acquire needle tip image data, and transmit it to the computer;

[0008] S3: The computer compares the needle tip image data with the broken needle image template to determine whether a broken needle has occurred;

[0009] S4: When the computer identifies a broken needle, it outputs a prompt message to the staff. Based on the above settings, the automatic detection of broken needles on the flat knitting machine is completed, so that the staff can be notified of the occurrence of broken needles and yarns in a timely manner, so that the staff can continue and handle the situation in a timely manner.

[0010] Preferably, the method for comparing the needle tip image data with the broken needle image template in S3 includes the following steps:

[0011] A1: Extract the pixel colors of the textile thread region in the broken needle image template to obtain the pixel color template;

[0012] A2: Compare the pixel colors of the textile thread area in the needle tip image data with the pixel color template;

[0013] A3: If the comparison fails, mark the needle tip image data as suspicious data;

[0014] A4: Artificial intelligence algorithms are used to perform secondary recognition and comparison on needle tip image data; preliminary screening is performed by directly checking whether the color of the textile thread appears in the shooting area, which can improve the efficiency of recognition and reduce the demand for computing power. Then, artificial intelligence algorithms are used for secondary recognition to improve accuracy.

[0015] Preferably, in S2, when a solid color plate is set facing the vision sensor and the needle tip image data is captured, the solid color plate serves as the background. By relying on the solid color plate as the background, the color difference between the needle tip area and the textile thread and the background can be improved, making it easier to identify.

[0016] Preferably, in step S2, a soft light supplement lamp is installed on the side of the vision sensor. The soft light supplement lamp illuminates the shooting area of ​​the vision sensor from the side. By setting the soft light supplement lamp, the shooting area can be illuminated, which helps to improve the clarity of the acquired pinhead image data.

[0017] Preferably, in S1, an area is defined based on the movable range of the knitting needle of the flat knitting machine. Within this area, jitter offset image templates are captured in intervals of 0.1 to 1 mm, and multiple jitter offset templates are uploaded to the computer. When the needle tip image data and the broken needle image template are not successfully compared, the computer performs a second comparison between the needle tip image data and the jitter offset template, thereby improving the accuracy of recognition and reducing false alarms.

[0018] Preferably, in step S4, the prompt information output by the computer includes needle tip image data and broken needle image template. This setting allows staff to make direct comparisons and judgments, making it easier to ascertain the specific situation.

[0019] Preferably, in A1, broken needle image templates under different weather times and seasons are collected, and then statistics are performed to define the color recognition range for the pixel color template based on the statistical data.

[0020] Preferably, the solid color plate is green.

[0021] The advantages of this invention are:

[0022] 1. This invention relies on a vision sensor to capture images of the knitting needles on a flat knitting machine, and then compares them with a pre-prepared image template of broken needles to automatically detect broken needles on the flat knitting machine. This allows workers to be aware of broken needles and yarns in a timely manner, so that they can promptly resume and handle the situation.

[0023] 2. This invention relies on setting up a soft light supplement lamp to provide supplemental lighting for the shooting area, which helps to improve the clarity of the acquired pinhead image data, and the side illumination can reduce the problem of reflection. Attached Figure Description

[0024] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1 This is a flowchart of the method of the present invention;

[0026] Figure 2 This is a flowchart of the method for comparing needle tip image data with broken needle image templates according to the present invention. Detailed Implementation

[0027] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0028] Specific implementation examples are given below.

[0029] Please see Figure 1 As shown, a method for detecting broken needles on a flat knitting machine based on a vision sensor includes the following steps:

[0030] S1: Take a pre-shot image template of a broken needle on a flat knitting machine and upload it to the computer;

[0031] S2: Use a vision sensor to photograph the needle tip of the flat knitting machine, acquire needle tip image data, and transmit it to the computer;

[0032] S3: The computer compares the needle tip image data with the broken needle image template to determine whether a broken needle has occurred;

[0033] S4: When the computer identifies a broken needle, it outputs a prompt message to the staff. During use, the computer relies on a vision sensor to capture images of the flat knitting machine needles and then compares them with a pre-prepared image template of a broken needle. This completes the automatic detection of broken needles on the flat knitting machine, allowing staff to be aware of the occurrence of broken needles or yarn in a timely manner, so that they can promptly reconnect and handle the situation.

[0034] For further details, please refer to Figure 2 The method for comparing needle tip image data with broken needle image templates in S3 includes the following steps:

[0035] A1: Extract the pixel colors of the textile thread region in the broken needle image template to obtain the pixel color template;

[0036] A2: Compare the pixel colors of the textile thread area in the needle tip image data with the pixel color template;

[0037] A3: If the comparison fails, mark the needle tip image data as suspicious data;

[0038] A4: Use artificial intelligence algorithms to perform secondary recognition and comparison on needle tip image data; when using it, if no needle breakage occurs, the needle tip area of ​​the flat knitting machine needle should have textile thread. Initial screening is performed by directly checking whether the color of the textile thread appears in the shooting area, which can improve the efficiency of recognition and reduce the demand for computing power. Then, a secondary recognition is performed using artificial intelligence algorithms to improve accuracy.

[0039] Furthermore, in S2, a solid color plate is set facing the vision sensor. When capturing and acquiring needle tip image data, the solid color plate serves as the background. In use, relying on the set solid color plate as the background can improve the color difference between the needle tip area and the textile thread and the background, making it easier to identify.

[0040] Furthermore, in S2, a soft light supplement lamp is installed on the side of the vision sensor, which illuminates the shooting area of ​​the vision sensor from the side. In use, by setting the soft light supplement lamp, the shooting area can be illuminated, which can improve the clarity of the acquired pinhead image data, and the side illumination can reduce the problem of reflection.

[0041] Furthermore, in S1, an area is defined based on the movable range of the flat knitting needle's vibration. Within this area, vibration offset image templates are captured in intervals of 0.1 to 1 mm, and multiple vibration offset templates are uploaded to the computer. When the needle tip image data fails to match the broken needle image template, the computer performs a secondary comparison between the needle tip image data and the vibration offset template. During use, the flat knitting needle operates at high speed, inevitably causing some vibration, which leads to a shift in the recognition area. This setting simulates the positional shift caused by the vibration of the flat knitting needle. Multiple vibration offset image templates are prepared in advance through pre-simulation of vibration, and then a secondary comparison is performed to improve the recognition accuracy and reduce false alarms.

[0042] Furthermore, in S4, the prompt information output by the computer includes needle tip image data and broken needle image template; when in use, this setting allows staff to easily make direct comparisons and judgments, and to easily ascertain the specific situation.

[0043] Furthermore, in A1, broken needle image templates under different weather conditions and seasons are collected and then statistically analyzed. Based on the statistical data, the color recognition range of the pixel color template is defined. In use, this setting can reduce the influence of external light on the color of the captured needle tip image data, making it easier to identify.

[0044] Furthermore, the solid color plate is green.

[0045] In the description of this specification, references to terms such as "an embodiment," "example," "specific example," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0046] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the claimed invention.

Claims

1. A method for detecting broken needles on a flat knitting machine based on a vision sensor, characterized in that: The method for detecting broken needles includes the following steps: S1: Take a pre-shot image template of a broken needle on a flat knitting machine and upload it to the computer; S2: Use a vision sensor to photograph the needle tip of the flat knitting machine, acquire needle tip image data, and transmit it to the computer; S3: The computer compares the needle tip image data with the broken needle image template to determine whether a broken needle has occurred; S4: When the computer identifies a broken needle, it outputs a prompt message to the staff. The method for comparing needle tip image data with broken needle image templates in S3 includes the following steps: A1: Extract the pixel colors of the textile thread region in the broken needle image template to obtain the pixel color template; A2: Compare the pixel colors of the textile thread area in the needle tip image data with the pixel color template; A3: If the comparison fails, mark the needle tip image data as suspicious data; A4: Use artificial intelligence algorithms to perform secondary recognition and comparison of needle tip image data; In S2, a solid color plate is set facing the vision sensor, and when capturing and acquiring pinhead image data, the solid color plate serves as the background. In S2, a soft light is installed on the side of the vision sensor, and the soft light illuminates the shooting area of ​​the vision sensor from the side. In S1, an area is defined based on the movable range of the knitting needle of the flat knitting machine. Within this area, shake offset image templates are captured in intervals of 0.1 to 1 mm, and multiple shake offset templates are uploaded to the computer. When the needle tip image data and the broken needle image template cannot be compared successfully, the computer performs a second comparison between the needle tip image data and the shake offset template. In step S4, the prompt information output by the computer includes needle tip image data and a broken needle image template. In A1, broken needle image templates under different weather times and seasons are collected, and then statistics are performed. Based on the statistical data, the color recognition range of the pixel color template is defined. The solid color panel is green.

Images