Quality control method for controlling dye homogeneity in dyeing machines

The quality control method addresses the lack of real-time feedback in dyeing machines by using camera-based image analysis and communication protocols to enhance dye homogeneity, achieving precise and rapid quality control with reduced poor-quality products.

WO2024242646A1PCT designated stage expired Publication Date: 2024-11-28KARABUDAK MUHENDISLIK MAKINA IMALAT SANAYI LTD STI
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
PCT/TR2023/050776
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Filing Date
2023-08-04
Publication Date
2024-11-28

AI Technical Summary

Technical Problem

Current quality control systems in dyeing machines lack digital automation, real-time feedback, and intelligent algorithms, resulting in low-quality products due to inadequate dye homogeneity control, which cannot be dynamically corrected during the dyeing process.

Method used

A quality control method using real-time image analysis and communication protocols to transmit machine parameters for improving dye homogeneity, involving camera-based image collection, processing, and transmission of correction data to the dyeing machine for continuous improvement.

Benefits of technology

Ensures high dye homogeneity with detection accuracy of ±0.01 mm and average quality control time of 120 milliseconds, reducing poor-quality products by enabling instantaneous machine parameter updates and dynamic corrections through Modbus and Ethercat protocols.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure TR2023050776_28112024_PF_FP_ABST
    Figure TR2023050776_28112024_PF_FP_ABST
Patent Text Reader

Abstract

The invention relates to a quality control method developed especially for use in the dyeing industry, which enables intelligent dyeing (e.g. especially in wood dyeing machines) and quality control processes using real-time detection methods, and which enables the improved machine parameters to be transmitted to the dyeing machines through communication protocols to increase the dye homogeneity in the dyeing machines.
Need to check novelty before this filing date? Find Prior Art

Description

[0001] QUALITY CONTROL METHOD FOR CONTROLLING DYE HOMOGENEITY IN DYEING MACHINES

[0002] Technical Field

[0003] The invention relates to a quality control method developed especially for use in the dyeing industry, which enables improved machine parameters to be transmitted to the machine through communication protocols to increase dye homogeneity in dyeing machines.

[0004] Prior Art

[0005] In the state of the art, there is no digital automation method in the units included in the quality control systems of the dyeing machines. Said methods do not provide the machines with feedbacks, notifications, corrections or fixes to improve the quality of the dye.

[0006] In other quality control units used in the state of the art, only the homogeneity of the dye applied to the parts is controlled. There are no systems (methods) with intelligent algorithms for improving the aforementioned control results. The result of not ensuring improvement is low quality products. In addition, since the operations to ensure homogeneity cannot be transferred to the machine, the parts dyed in the process are only analyzed afterwards, and the quality control units cannot communicate with the machine instantaneously and make corrections dynamically after each part.

[0007] There is a need to develop new systems to eliminate the disadvantages mentioned above, which are / are not present in the state of the art.

[0008] Summary of the Invention

[0009] The object of the invention is to develop a quality control method which enables intelligent dyeing (e.g. especially in wood dyeing machines) and quality control processes using real-time detection methods, and which enables the machine parameters to be sent to the dyeing machines through communication protocols to increase the dye homogeneity in the dyeing machines.

[0010] Another object of the invention is to analyze, save and report the dye homogeneity of the product dyed in the dyeing machines, and to communicate with the dyeing machine to transfer the operations and corrections to increase the measured dye homogeneity.

[0011] Yet another object of the invention is to make the dyeing machines more intelligent and to enable the said machines to continuously receive correction information by means of the quality control unit and method in order to ensure that the dye layer provided on the products is more homogeneous.

[0012] Another object of the invention is to reduce the number of products with poor dye quality that are dyed with the dyeing machines.

[0013] Another object of the invention is to ensure that the quality controls of the dyed products are carried out in a fast and easy way and to report them digitally.

[0014] Description of the Drawings

[0015] Fig. 1. A representative perspective view of the quality control unit according to the invention.

[0016] Fig. 2. A representative perspective view of the integration of the quality control unit according to the invention and the dyeing machine conveyor.

[0017] Fig. 3. A representative general flow chart of the method according to the invention.

[0018] Description of the Reference Numbers in the Drawings

[0019] For a better understanding of the invention, the corresponding reference numbers in the drawings are given below;

[0020] 1. Led Lighting

[0021] U. Unit

[0022] 101 . Parts coming out of the dyeing machine entering the quality control unit 102. At least one camera being triggered by the entry of the parts and starting to collect images

[0023] 103. Saving the collected images in RAM memory with at least one storage unit

[0024] 104. Transferring the images taken by the camera to at least one device via at least one communication unit after the part passes completely under the quality control unit and making geometric and / or spectral arrangements

[0025] 105. Applying filters to the collected images

[0026] 106. Merging the images as a result of pre-processing and running the object recognition algorithm on the merged image

[0027] 107. Transforming the merged image into a gray-scale sample space in at least one database

[0028] 108. Taking an average of the pixel projection values

[0029] 109. Obtaining the difference between the pixel projection values and the calculated average

[0030] 110. Creating a map of dye homogeneity

[0031] 111. Determining the dyeing improvements to be made according to the dye homogeneity map

[0032] 112. Transmitting the dyeing improvements to the machine via a communication unit

[0033] Detailed Description of the Invention

[0034] The exemplary embodiments are described in more detail below with reference to the accompanying descriptions. However, the embodiments may take different forms and should not be construed as limited to the embodiments set out here. Instead, these exemplary embodiments are provided to ensure that this disclosure is comprehensive and its scope is fully communicated to those skilled in the art.

[0035] The terminology used in this description is intended to describe the specific exemplary embodiments only and is not intended to be limiting. As used herein, the forms "a", "an", "at least" and "preferably" are intended to include the plural forms unless the context clearly indicates otherwise. The terms "comprising" and / or "including", when used in this specification, includes the specified features, integers, steps, operations, elements and / or components, but does not preclude the existence or addition of one or more other features, integers, steps, operations, elements and / or components. The invention is a quality control method developed especially for use in the dyeing industry, which enables the improved machine parameters in the quality control unit (U) to be transmitted to the machine through communication protocols to increase dye homogeneity in dyeing machines, comprising the steps of:

[0036] Parts coming out of the dyeing machine entering the quality control unit (U) (101 )

[0037] At least one camera being triggered by the entry of the parts and starting to collect images (102)

[0038] Saving the collected images in RAM memory with at least one storage unit

[0039] (103)

[0040] Transferring the images taken by the camera to at least one device via at least one communication unit after the part passes completely under the quality control unit (U) and making geometric and / or spectral arrangements

[0041] (104)

[0042] Applying filters to the collected images (105)

[0043] Merging the images as a result of pre-processing with geometric and spectral corrections and running the object recognition algorithm on the merged image (106)

[0044] Transforming the merged image into a gray-scale sample space in at least one database (107)

[0045] Taking an average of the pixel projection values (108)

[0046] Obtaining the difference between the pixel projection values and the calculated average (109)

[0047] Creating a map of dye homogeneity (110)

[0048] Determining the dyeing improvements to be made according to the dye homogeneity map (11 1 )

[0049] T ransmitting the dyeing improvements to the machine via a communication unit (112).

[0050] In step 102, the white balance and exposure parameters of the camera are fixed with at least one interface to ensure that images are collected under equivalent conditions. Said interface is a software language and Pyhton is preferred in an exemplary embodiment of the invention, but the embodiment is not limited thereto. In a preferred embodiment of the invention, at least one digital camera is used, which is 13MP with a 90° diagonal viewing angle, fixed-focus, adjustable white balance and exposure time, but the embodiment is not limited thereto.

[0051] In step 103, regardless of the part size, an image is collected for every 60 millimeters of the part as it moves continuously on the conveyor positioned below the quality control unit (U). The images collected along the parts are stored in the RAM memory of at least one device, which is a computer, for fast processing with a digital image processing interface.

[0052] The invention comprises at least one dimmable led lighting (1 ), which ensures that the right, left and middle regions of the part flowing on the conveyor are homogeneously illuminated, which is positioned in at least said three regions, and the light intensity of which can be adjusted. In the images collected in step 103, sufficient light conditions are provided by the led lighting (1 ).

[0053] In step 104, the images collected in the pre-processing stage are re-sampled with the Nearest Neighbor method, which gives fast results, the resolution is reduced to 640x480 for faster processing on the image, and geometric and / or spectral corrections are made.

[0054] In step 105, a Gaussian filter with a Kernel size of 5x5 is applied to remove noise from the image. A small Kernel size of 5x5 was preferred as the images were collected under sufficient light. Noise in the image is removed with a Gaussian filter.

[0055] In step 106, as the part moves along the conveyor, the dynamically collected images are merged as a result of pre-processing to create an image of the entire part. In said resulting image, the dyed part is detected and separated from the background with the object recognition algorithm developed using the latest Y0I0V6 algorithm, which has a fast learning method and short detection time.

[0056] In step 106, after obtaining the coordinates of the edge points of the part, a polygon is defined in the pixel coordinate system in which these points are formed. If the object area is larger than the area specified in the interface (106a), the process continues with step 107. If the object area is not larger than the area specified in the interface, it is assumed that the captured part is too small to analyze and a return to step 101 is made with the command to wait for the next part (106b). Said stage also helps to eliminate part recognition errors caused by light changes that may occur due to sensitivity on the detection side.

[0057] The edge coordinates of the part obtained with step 106 are compared with the edge coordinates collected with at least one lidar sensor. The accuracy of the process is confirmed with the said lidar sensor.

[0058] In steps 108 and 109, for the part with previously detected corner points in the gray-scale color space image, the pixel projection values within the polygon formed by these points are averaged. Since the color homogeneity on the object is examined, only the pixel projection values covered by the detected polygon are taken into account when averaging. A zero matrix of the same size (resolution) as the merged image is created. In the gray-scale color space image, for the object with previously detected corner points, the difference between the projection values of the pixels within the polygon formed by these points and the average projection values calculated in step 109 is obtained.

[0059] In step 110, the result of the projection value difference is saved in the corresponding position of the blank matrix (all pixel values are 0), which is created to represent a blank image, in order to produce the heat map. As a result of this step, a map is created showing the difference of each pixel representing the part surface from the average projection value of the part.

[0060] In step 11 1 , the under-dyed or over-dyed regions that disrupt the homogeneity in the obtained dye homogeneity map are detected by comparing with the user-entered limit value in the interface (values above the user-entered limit value).

[0061] In step 112, in the areas that are above the maximum value or below the minimum value saved in the interface and in the regions where more dye needs to be applied to ensure dye homogeneity on the part, the information about increasing the flow rate to the spray gun together with the coordinates representing the corresponding regions or, if the flow rate to the spray gun is constant, the dyeing route information about covering the corresponding region and allowing the dyeing machine to apply one or more layers of dye is transmitted to the machine. With the method according to the invention, the quality control units (U) positioned at the inlet and outlet of the dyeing machine and the state of the part to be dyed before and after dyeing can be detected and compared. As a result of the said comparison, it is possible to measure what kind of result (provided dye homogeneity) will be obtained when the parts, which have not yet been dyed and are entering the machine to be dyed, are dyed with the current dyeing parameters.

[0062] With the quality control method according to the invention, it is ensured that the machine parameters are continuously updated instantaneously according to the results of the dye homogeneity analysis of the parts that have previously entered the machine and dyed in order to ensure that the parts that have just entered the machine come out with a homogeneous dye.

[0063] With the invention, a detection accuracy of ± 0.01 mm and an average quality control time of 120 milliseconds per part are achieved. With the method according to the invention (in particular step 102), the capacity to collect an image every 30 milliseconds is achieved.

[0064] The data mentioned in the method steps such as object coordinates, numeric data representing homogeneity, coordinates to be followed for dyeing routes and flow rate values can be transferred to the machine using Modbus and / or Ethercat communication protocols with a communication delay of 1 millisecond. Therefore, according to the quality control results of the products that will enter the machine, it is ensured that the regions where more or less dye should be applied are dyed accordingly.

[0065] In summary, in the method according to the invention, when any part starts to pass under the quality control unit (U), the camera that the parts pass over recognizes that the background sheet is covered by the part by obtaining the difference of pixels and detects that the part starts to pass. After the built-in camera in the quality control unit (U) detects the part entering the unit (U), it starts to dynamically collect images of the entering parts as they flow on the conveyor belt.

[0066] Industrial Applicability of the Invention The invention relates to a quality control method developed especially for use in the dyeing industry, which enables improved machine parameters to be transmitted to the machine through communication protocols to increase dye homogeneity in dyeing machines, and is applicable to the industry.

[0067] The invention is not limited to the foregoing description, and one skilled in the art can easily come up with different embodiments of the invention. These should be considered within the scope of the protection sought by the claims of the invention.

Claims

CLAIMS1 . A method developed especially for use in the dyeing industry, which enables the improved machine parameters in the quality control unit (U) to be transmitted to the machine through communication protocols to increase dye homogeneity in dyeing machines, characterized in that it comprises the steps of:Parts coming out of the dyeing machine entering the quality control unit (U) (101 )At least one camera being triggered by the entry of the parts and starting to collect images (102)Saving the collected images in RAM memory with at least one storage unit(103)Transferring the images taken by the camera to at least one device via at least one communication unit after the part passes completely under the quality control unit (U) and making geometric and / or spectral arrangements(104)Applying filters to the collected images (105)Merging the images as a result of pre-processing with geometric and spectral corrections and running the object recognition algorithm on the merged image (106)Transforming the merged image into a gray-scale sample space in at least one database (107)Taking an average of the pixel projection values (108)Obtaining the difference between the pixel projection values and the calculated average (109)Creating a map of dye homogeneity (110)Determining the dyeing improvements to be made according to the dye homogeneity map (11 1 )T ransmitting the dyeing improvements to the machine via a communication unit (112).

2. A method according to claim 1 , characterized in that in step 102, the white balance and exposure parameters of the camera are fixed with at least one interface to ensure that the images are collected under equivalent conditions.

3. A method according to claim 2, characterized in that it comprises at least one digital camera with a diagonal viewing angle, fixed-focus, and adjustable white balance and exposure time.

4. A method according to claim 1 , characterized in that in step 103, the digital image is stored in the RAM memory of at least one device, which is a computer, for fast processing with a processing interface.

5. A method according to claim 4, characterized in that it comprises at least one dimmable led lighting (1 ), which ensures that the right, left and middle regions of the part flowing on the conveyor are homogeneously illuminated, which is positioned in at least said three regions, and the light intensity of which can be adjusted.

6. A method according to claim 5, characterized in that in step 105, a Gaussian filter with a Kernel size of 5x5 is applied to remove noise from the image.

7. A method according to claim 6, characterized in that in step 106, after obtaining the coordinates of the edge points of the part, a polygon is defined in the pixel coordinate system in which the said edge points are formed.

8. A method according to claim 7, characterized in that in step 106, if the object area is larger than the area specified in the interface (106a), the process continues with the step (107) of transforming the merged image into a gray-scale sample space in at least one database.

9. A method according to claim 8, characterized in that in step 106, if the object area is not larger than the area specified in the interface, it is assumed that the captured part is too small to analyze and a return to the step (101 ) of entry of the parts coming out of the dyeing machine to the quality control unit (U) is made with the command to wait for the next part (106b).

10. A method according to claim 9, characterized in that in step 106, the edge coordinates of the obtained part are compared with the edge coordinates collected with at least one lidar sensor. 1 1 . A method according to claim 10, characterized in that in steps 108 and 109, for the part with detected corner points in the gray-scale color space image, the pixel projection values within the polygon formed by these points are averaged.

12. A method according to claim 1 1 , characterized in that in step 1 10, the result of the projection value difference is saved in a position of the blank matrix with all pixel values of 0 created to represent a blank image, in order to produce the heat map.

13. A method according to claim 12, characterized in that in step 1 11 , the under- dyed or over-dyed regions that disrupt the homogeneity in the obtained dye homogeneity map are detected by comparing with the user-entered limit value in the interface.

Citation Information

Patent Citations

  • Artificial board surface defect detection device and method

    CN111612737A

  • Imaging system for analysis of wood products

    US10825164B1

  • Method and system for detecting the quality of debarking at the surface of a wooden log

    US20130333805A1

  • Method and apparatus for quantifying pigment dispersion quality by paint drawdown

    WO2007067801A2