MACHINE AND METHOD TO CONTROL TEXTILE QUALITY

Abstract

Machine and method for controlling textile fabric defects during the textile production and method for retrofitting a circular knitting machine having a fixed support structure, a rotational support structure, said system comprising: a digital camera for capturing digital images of the knitted textile fabric, a data processor for processing the captured digital images, a camera support structure for holding the camera, a lighting system to illuminate the knitted textile fabric from the camera side for capture by the digital camera; wherein the camera and back support structures are fixed to the rotational structure.

Claims

1. A circular knitting machine comprising a fixed support structure, a rotational support structure, and a system for controlling textile fabric defects, said circular knitting machine comprising: a roller for flattening a circular knitted web into said textile fabric, and a cylinder arranged on a lower part of said rotational structure for rolling-in the flattened textile fabric; said system comprising: a digital camera for capturing digital images of the knitted textile fabric, a data processor for processing the captured digital images, and a camera support structure for holding the camera, wherein the camera support structure is fixed to the rotational structure.

2. The circular knitting machine according to claim 1, wherein said camera support structure is arranged such that the digital camera captures digital images of the flattened textile fabric.

3. The circular knitting machine according to claim 1 wherein the data processor is arranged for: processing the captured digital images for detecting textile defects in real time, in particular for stopping the motion of the circular knitting machine for preventing more defects.

4. The circular knitting machine according to claim 1, wherein said digital camera is a one-dimensional camera or a 2D camera.

5. The circular knitting machine according to claim 1, wherein said camera support structure comprises a front light for illuminating the knitted textile fabric.sub.s from the camera side.

6. The circular knitting machine according to claim 1, further comprising an additional support structure and a back light arranged on said additional support structure.sub.s for illuminating the knitted textile fabric from a side opposite camera side and fixed to the rotational structure, wherein said back light comprises a light source for illuminating the knitted textile fabric.

7. The circular knitting machine according to claim 1, wherein said circular knitting machine comprises: a front light arranged on said camera support structure to illuminate the knitted textile fabric from the camera side for capture by the digital camera, an additional support structure and a back light arranged on said additional support structure to illuminate the knitted textile fabric from a side of the textile fabric opposite to the camera for capture by the digital camera, wherein the camera support structure and the additional support structure are fixed to the rotational structure.

8. The circular knitting machine according to claim 1 wherein the camera is arranged to accommodate an optical device for enhancing resolution of the image captured from knitted textile.

9. The circular knitting machine according to claim 8, wherein the optical device comprises one or more optical lenses arranged for signal acquisition.

10. The circular knitting machine according to claim 9, wherein the optical lenses are selected for predetermined distances for obtaining an imaging resolution of less than one millimetre.

11. The circular knitting machine according to claim 1 wherein the front light and back light sources comprise light sources with different wavelengths selected from the group consisting of: infrared light, visible light, ultraviolet light, and combinations thereof.

12. A method for obtaining a retrofitted circular knitting machine having a fixed support structure, a rotational support structure, with a system for controlling textile fabric defects, the method comprising: fitting said circular knitting machine with the system described in claim 1 to obtain the retrofitted circular knitting machine.

13. A method for controlling textile fabric defects of the circular knitting machine according to claim 1, comprising: capturing digital images of the knitted textile fabric, and using the data processor for processing the captured digital images for detecting knitted textile fabric defects.

14. The method for controlling textile fabric defects in the circular knitting machine according to claim 13, wherein said camera support structure is arranged such that the digital camera captures digital images of the flattened textile fabric.

15. The method according to claim 13, wherein the capturing digital images of the knitted textile fabric is made between the roller and the cylinder with the digital camera.

16. The method according to claim 13, wherein the capturing digital images of the knitted textile fabric is made in rotational synchronisation between the camera and the knitted textile fabric.

17. The method according to claim 13, wherein the method further comprises using a front light to illuminate the knitted textile fabric from the camera side for capture by the digital camera, and using a back light to illuminate the knitted textile fabric from the side opposite to the camera for capture by the digital camera.

18. The method according to claim 13, further comprising using the data processor for processing the captured digital images in real time for detecting textile defect.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0087] The following figures provide preferred embodiments for illustrating the disclosure and should not be seen as limiting the scope of disclosure.

[0088] FIG. 1: Schematic conceptual representation of an embodiment of a circular (or Weft) knitting machine having a system according to the disclosure fixed to lateral supports of the rotational part of the machine.

[0089] FIG. 2: Schematic representation of an embodiment of a data flow chart with a general view of the disclosed system.

[0090] FIG. 3: Schematic representation of the main elements of a main embodiment with front and back lighting system.

[0091] FIG. 4: Schematic representation of the main elements of a main embodiment with represented axis of rotation.

[0092] FIG. 5: Schematic representation of the perspective view of main elements of a main embodiment.

[0093] FIG. 6: Schematic representation of another perspective view of main elements of a main embodiment.

[0094] FIG. 7: Schematic representation of the main elements of the rotative structure (tubular fabrics being rolled) of an embodiment.

[0095] FIG. 8: Schematic representation of the main elements of the rotative structure (tubular fabrics being rolled) with alternative area to film according to an embodiment.

[0096] FIG. 9: Schematic representation of the main elements of the rotative structure (tubular fabrics being rolled) with alternative area to film according to an embodiment.

[0097] FIG. 10: Schematic representation of the main elements of the rotative structure (tubular fabrics being rolled) with alternative area to film according to an embodiment using reflective systems, such as mirrors, to acquire images in difficult areas.

DETAILED DESCRIPTION

[0098] The system according to an embodiment of the present disclosure is typically comprised by:

[0099] Camera (or alternative device to acquire a 2D or 1D signal from the textile surface. In this case: images). It is the data acquisition system, should be preferably able to acquire images on industrial conditions such as moving textile (depending on defect types, it is convenient to use a camera sensor with sensitivity on infrared and/or ultraviolet light and/or visible light), should be suitable in size and convenient to plug in the disclosed structures.

[0100] Optical system: Depending on the position as well as the distance between camera and fabrics, it is preferable to have the best optical system to enhance image resolution. In a first approach, this system is preferably composed by lenses that are possible to be regulated for better 2d or 1d signal acquisition.

[0101] Light source: The lighting system is preferential but could be optional. Depending on surrounding conditions and sufficient background illumination, it is possible to use the disclosed system with no back or front light. This also means that front and back structures are also possible to be removed. So as an alternative, the disclosed system could be used with front light only (tangential or not), backlight only, or none of them. It could be individual light or continuous, placed on one or both sides of the textile, with light normally incident or low-angle light. Each position enhances different properties and flaws, as well as highlight the textile structure and texture. The intensity should preferably be adjusted to the setup, in order to maximize image resolution and quality. Different wavelengths such as infrared, visible, and other light wavelengths) are also used to get different information about colour, texture, and material properties.

[0102] Processing unit: Hardware unit that receives the acquired data, processes it (using the disclosed software for 2d or 1d signal analysis) and acts depending on the result. It could communicate with the exterior.

[0103] Structure to hold components: The acquisition element and light sources should preferably be placed in specific positions, respecting distances, angles, and field of view. In order to guarantee such conditions, the present system is provided with a structural element that supports the other devices.

[0104] Computer server (optional): The disclosed system can preferably be communicating with the exterior, and there are several additional features that can be integrated by using external elements such as a server. One preferable feature is the real-time monitoring from phones or computers, where an app could show in real time the machine's production with images and quality reports. This system could be physically outside the detection device, or not.

[0105] This disclosure aims to determine if there is a defective product or not. Despite that, it is preferably possible to distinguish between all different defects and save information for reporting and statistics.

[0106] The variables that affect this system are:

[0107] Acquisition system: 2d signal acquisition using a camera or alternative. It should preferably present a low time exposure (fast shutter) in order to take pictures in an hostile environment (features described previously);

[0108] Camera position (position, distance, and angle): The acquisition system must preferably be wisely placed in order to get more and complete information about the textile structure and properties. It could be placed on the fixed part (top of the machine) inspecting fabrics right next to the needles. It could also be installed on the bottom part of the machine (moving or not). In case the camera is fixed (not rotating), it should be prepared to capture images in synchrony with the machine movement. In case that it is placed on the moving part, the camera is installed as FIG. 4 suggests. With this setup, all the structure is preferably connected to the rotating part, and as consequence, it is rotating too. The camera could be placed at several angles and distances, only restricted by the machine structures;

[0109] Lighting (position and angle): The lighting system is also presented in FIG. 4. It could be composed by singular light sources, or continuous. Different angles and distances could be used, the present system allows to change it manually. Front low-angle illumination is preferably used. Depending on machine type, the light system could be used in different. Light sources on the same side of the camera (front light) are used to detect some defects (mainly shape flaws). Light sources on the opposite side of the camera (backlight) are used to detect other defects and properties (mainly structural flaws) and could be placed in several distances and angles. The present disclosure takes into account both approaches (simultaneously or not). Regarding light wavelength, the disclosed system is prepared to work with visible light spectrum as well as near-infrared light. Alternative wavelengths (such as UV or infrared) could be used as well, which are useful for detecting problems in coloured fabrics;

[0110] Optical system: Depending on the camera-textile distance, as well as the textile type (mainly on the density of yarns), an optical system must preferably be used in order to allow the camera to acquire a complete 2d signal without losing information. The present disclosure takes into account optical lenses that are preferably chosen for specific distances, and preferably guarantee imaging resolution of less than one millimeter, where for example for Lycra (™) materials the defect may have 0.5 mm;

[0111] Variables for thresholding: There are several types of fabrics that could be produced: Jersey, Interlock, RIB, American fleece, etc. Each one has a different structure that needs to be considered in the disclosed system. For defect detection, the inspection software must preferably process the signal and check if all image parameters are acceptable. The threshold variables should preferably be different for different types of textile. These variables are automatically controlled preferably by the system itself, by machine learning methods or simply by variance calculation and adjustment. of settings. Textile materials are not perfectly uniform because of its structure, hairiness and fibres. For that reason, uniformity analysis must preferably take into account uniformity error margins that could be dynamically changing or not;

[0112] Types of faults: Depending on the defect type, different algorithms will detect it. The present disclosure preferably follows a certain order to efficiently check the production quality. By computing Gaussian filters, Local Binary Pattern (LBP) algorithms, machine learning techniques and others, it is possible to identify patterns and non-uniformities on 2D signals. It follows that each defect is associated with a certain type of pattern or non-uniformity, and consequently it is possible to register each flaw individually.

[0113] The present disclosure hardware is preferably responsible for:

[0114] Signal acquisition; [0115] Controlled environment regarding lighting conditions; [0116] Data communication between components (camera, processing unit, server, etc) Physical protection and mechanical supporting; [0117] Actuating on machinery by stopping it or activating an alert system, or both.

[0118] Regarding software, it is preferably responsible for: [0119] Processing and analysing acquired data (images, light sources, machine status) [0120] Control the communication between components; [0121] Control hardware for actuating on machinery; [0122] Data analysis for statistics and reporting.

[0123] There are alternative ways of using the disclosed system and method: [0124] Use multiple cameras for image acquisition on full textile width. This could be done by supporting multiple cameras on the structure placed in the rotative part of the knitting machine. One example of this approach is depicted on FIGS. 1, 3, 4, 5 and 6; [0125] Use linear cameras (1D signal) synchronized with the textile movement in order to build a 2d signal. This alternative will lead to a 2d signal (image or image sequences) as the disclosure suggests; [0126] It is also possible to integrate it in several positions such as the cylinder, doors, fixed structure on the bottom of the machine, etc. All alternatives have in common the use of image acquisition for defect detection and automatic quality control of knitted fabrics in real time; [0127] It is also possible to integrate this system in other machines such as different circular knitting machinery (different brands or models), inspection machinery or dye-house machinery. [0128] All positions have in common the use of 2d signal acquisition and processing for automatic quality control of knitted fabrics in real time. [0129] Alternative methods for communication between elements could be used. Acquired data from the camera to the processing unit (wireless or not). Communication from the acquisition element to exterior could also be done using different wireless communication systems. [0130] Light sources could be used with a front light arranged on said frontal support structure to illuminate the knitted textile fabric from the camera side for capture by the digital camera, and a back light arranged on said back support structure to illuminate the knitted textile fabric, from a side of the textile fabric opposite to the camera, for capture by the digital camera; wherein the frontal and back support structures fixed to the rotational structure. [0131] Alternative illumination systems could be used. Changing the light angle, the number of light sources, distance to fabrics, will lead to different image properties. However, the disclosed system is prepared to work properly with different light, adjust its variables for several lighting conditions. The acquisition module could use alternative wavelengths such as infrared, ultraviolet, or specific visible light (front light as well as backlight). [0132] If the rotative part of the knitting machine does not allow to place the structures described before, it is possible to use additional structures and systems to signal acquisition. If there is no space for camera positioning, it is possible to use reflection systems, such as mirrors, to give flexibility to the disclosed system. This is particularly useful for situations where the textile fabrics do not allow to place cameras in a perpendicular angle. An example is depicted in FIG. 10.

[0133] The term “comprising” whenever used in this document is intended to indicate the presence of stated features, integers, steps, components, but not to preclude the presence or addition of one or more other features, integers, steps, components or groups thereof. The disclosure should not be seen in any way restricted to the embodiments described and a person with ordinary skill in the art will foresee many possibilities to modifications thereof. The above described embodiments are combinable. The following claims further set out particular embodiments of the disclosure.