Ultrasonic SEAM inspection system and method comprising an image processing algorithm

Abstract

The invention relates to a system and method for determining the seam quality by projecting light at an angle onto fabric subjected to ultrasonic welding to create shadows caused by depth on the fabric, calculating the area of the generated shadows, and comparing the obtained values with reference values.

Description

[0001] ULTRASONIC SEAM INSPECTION SYSTEM AND METHOD COMPRISING AN

[0002] IMAGE PROCESSING ALGORITHM

[0003] TECHNICAL FIELD

[0004] The invention relates to an ultrasonic seam inspection system and method comprising an image processing algorithm, for inspecting the seam quality of fabric subjected to ultrasonic welding.

[0005] The invention relates to a system and method that determine the seam quality by projecting light onto the fabric subjected to ultrasonic welding at an angle to create shadows caused by depth on the fabric, calculating the shadow area formed, and comparing the obtained values with reference values.

[0006] PRIOR ART

[0007] Ultrasonic welding is a welding method that utilizes high-frequency mechanical vibrations. The ultrasonic welding method is one of the alternatives to conventional sewing techniques used for the fast and clean joining of different textile surfaces. Compared to traditional sewing methods, ultrasonic welding offers various advantages such as a much shorter stitching and welding time and greater cost efficiency.

[0008] Ultrasonic refers to sound waves with frequencies above the human hearing range, typically around or above 20 kHz. The principle of ultrasonic bonding is based on the use of high-frequency vibration to join materials with thermoplastic structures. As a result of the vibration, heat is generated within the material, causing it to melt, fuse, and form a new bond. Ultrasonic welding is considered an energy-saving method and, since it does not require materials such as thread used in conventional sewing methods, it is also regarded as a more easily recyclable process. The ultrasonic bonding / welding method is utilized in various industries such as automotive, textile, apparel, medical, packaging, electronics, and filtration. In the existing art, there are methods used to control the seam quality in the ultrasonic welding region. In these methods, the area is calculated directly. Application No. KR101305255B1 is an example of systems employing this method. In the present invention, it is mentioned that image processing and area calculation are used to analyze the quality of the ultrasonic welded region between tabs and terminals in battery modules or battery packs. However, in this patent application, only the boundary lines of the weld are calculated. Therefore, a reliable quality control cannot be achieved. Moreover, the method used performs quality control in a limited industrial sector.

[0009] In U.S. Patent No. US2013011717A1 , camera images are used for outer edge detection during ultrasonic welding and for sending corresponding signals to the controller.

[0010] Another U.S. Patent, No. US2022379410A1 , is integrated with an ultrasonic welding device and is used to monitor the horn and the welded area in real time.

[0011] As a result, the problems mentioned above, which cannot be resolved in light of the existing technology, have made it necessary to introduce an innovation in the related technical field.

[0012] SUMMARY OF THE INVENTION

[0013] The present invention relates to an ultrasonic welding control system comprising an image processing algorithm, which calculates the area of shadowed regions to determine the depth of the ultrasonic weld, thereby eliminating the disadvantages mentioned above and providing new advantages to the related technical field.

[0014] The main objective of the invention is to provide a control system and method for monitoring the seam quality in the ultrasonic welding region.

[0015] Another objective of the invention is to provide a system and method capable of determining how deep the weld has been made in the ultrasonic welding region, thereby enabling the assessment of its strength. The present invention, in order to achieve all the objectives mentioned above and those that will become apparent from the detailed description below, relates to an ultrasonic welding control system that controls the seam quality of the fabric subjected to ultrasonic welding by means of an image processing algorithm. The system comprises at least one camera positioned to capture video images of the fabric being processed by imaging the ultrasonic welding region, at least one light source positioned at a desired angle to create shadows on the fabric resulting from depth variations, and a computer that analyzes the images of the shadowed fabric captured by the camera through an embedded image processing algorithm. The computer calculates the total shadow area (AT) and determines the welding quality by comparing the calculated shadow area (AT) with predefined reference area parameters (ai, 32, as, 34) stored in the system database to identify the welding quality levels. The reference area parameters defining the welding quality levels may be increased or decreased as desired or as needed. These parameter values are predefined according to known quality levels and are used within the algorithm.

[0016] In comparison with Patent No. KR101305255B1 , in the present invention, shadows are formed by lights positioned at an angle along the production line, and deeper areas in the ultrasonic welded region produce more shadows. The shadow areas on the fabric are then calculated. In the Korean patent, however, only the boundary lines of the weld are calculated; therefore, it is not possible to analyze how deep the weld has been made and, consequently, its strength.

[0017] In comparison with Patent No. US2022379410A1 , in the present invention, the camera is not located at the ultrasonic welding site but positioned further along the production line, and it is used solely to inspect the weld seams.

[0018] In order to achieve all the objectives mentioned above and those that will become apparent from the detailed description below, the present invention is an ultrasonic welding control method that controls the seam quality of the fabric subjected to ultrasonic welding by means of an image processing algorithm, and comprises the following steps: Projecting light onto the fabric subjected to ultrasonic welding at an angle to create shadows on the fabric caused by depth, Capturing video images of the shadowed fabric, Creating image frames from the recorded videos,

[0019] Converting the format of the image frame for processing,

[0020] Blurring the image frame whose format has been changed,

[0021] Masking the image within a desired two-color range to make the shadows in the blurred image more distinct,

[0022] Defining the region to be analyzed for the masked shadows,

[0023] Detecting and marking the boundaries of the shadows within the defined region,

[0024] Calculating the area of the marked shadows,

[0025] Comparing the calculated shadow area (AT) with predefined seam quality reference area parameters (ai , 32, as, 34) and determining the seam quality.

[0026] In order to best understand the structure of the present invention and its advantages together with the additional components, it should be evaluated together with the figures described below.

[0027] BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 is a schematic view of the ultrasonic seam inspection system of the present invention.

[0029] Figure 2 is another schematic view of the ultrasonic seam inspection system of the present invention.

[0030] Figure 3 is yet another schematic view of the ultrasonic seam inspection system.

[0031] Figure 4 is a representative view of the shadowed fabric mentioned in the invention.

[0032] Figure 5 is a flow diagram representing the method of the invention. REFERENCE NUMERALS

[0033] 1 Camera

[0034] 2 Light Source

[0035] 3 Ultrasonic Weld Area

[0036] 4 Lightproof Protective Enclosure

[0037] 5 Computer

[0038] 6 Relay Switch

[0039] 7 Power Supply

[0040] G Shadow

[0041] DETAILED DESCRIPTION OF THE INVENTION

[0042] In this detailed description, the innovation of the invention is explained only with examples that do not impose any limiting effect, for better understanding of the subject matter. The invention relates to an ultrasonic welding control system and a control method comprising an image processing algorithm for detecting the seam quality of fabric subjected to ultrasonic welding.

[0043] Figure 1 is a schematic view of the ultrasonic welding control system of the invention. Accordingly, the ultrasonic welding control system comprises at least one camera (1 ) imaging the ultrasonic weld area (3), at least one light source (2) positioned at an angle relative to the ultrasonic weld area (3), a relay switch (6), and a power supply (7). In this system, in order to prevent the shadow (G) detection part of the algorithm from being affected by ambient light intensity and light color variations, the camera (1 ) and light source (2) are also positioned inside a lightproof protective enclosure (4), as shown in Figures 2 and 3. If the system is to be used without the lightproof protective enclosure (4), shadow (G) detection can be performed using an artificial intelligence-based or advanced image processing-based algorithm. The system also comprises a computer (5) in which the image processing algorithm is executed.

[0044] Light is projected at an angle onto the fabric processed in the ultrasonic weld area (3) by the light source (2). The position and orientation angle of the light source (2) can be adjusted automatically or manually depending on the size of the weld depth to be detected. For this purpose, the light source (2) can be connected very simply to a suitable location on the line, and its orientation can also be adjusted via a motorized or manual mechanical system in which it is mounted. This lighting creates shadows (G) on the fabric caused by depth variations. Figure 4 shows a representative view of the shadowed fabric. Video images of the shadowed fabric are captured by the camera (1 ). These video images are transferred to the computer (5) for processing. First, the computer (5) processor generates image frames from the video images (for example, in 1024x768 pixel RGB format). Then, the format of the image frame is converted to a suitable format for analysis (for example, from RGB format to HSV format). The format-converted image frame is blurred, and in the blurred image, shadows (G) are formed in regions with depth as a result of angled light reflection at the edges of the weld lines. To enhance these shadows (G), the image is masked within a desired two-color range. The boundaries of the area to be analyzed for the masked shadows (G) are defined. The shadows (G) within the defined area are detected and their boundaries are marked by the computer (5) processor. The area of the marked shadows (G) is calculated, and the total shadow area (AT) is compared with predefined reference area parameters (ai, 32, as, 34) to determine the quality.

[0045] Within the scope of the ultrasonic welding control method of the invention, video images of the shadowed fabric obtained through the camera (1) and light source (2) are processed by an image processing algorithm executed on the computer (5) to determine the seam quality of the fabric subjected to ultrasonic welding. The method comprises the following steps:

[0046] • Projecting light onto the fabric subjected to ultrasonic welding at an angle to create shadows (G) on the fabric caused by depth,

[0047] • Capturing video images of the shadowed fabric,

[0048] • Creating image frames from the recorded videos,

[0049] • Converting the format of the image frame for processing,

[0050] • Blurring the image frame whose format has been changed,

[0051] • Masking the image within a desired two-color range to enhance the visibility of the shadows (G) in the blurred image,

[0052] • Defining the region to be analyzed for the masked shadows (G), Detecting and marking the boundaries of the shadows (G) within the defined region,

[0053] Calculating the area of the marked shadows (G),

[0054] Comparing the calculated total shadow area (AT) with predefined seam quality reference area parameters (ai, 32, as, 34) stored in the system database and determining the seam quality.

[0055] As a result of comparing the calculated total shadow (G) area (AT) with the predefined reference area parameters (ai, 32, as, 34), the following conclusions regarding seam quality are reached:

[0056] AT < ai Very poor seam ai < AT < 32 Poor seam

[0057] 32 AT < as — >■ Medium-quality seam

[0058] 83 < AT < 34 Good seam

[0059] 34 AT Very good seam

[0060] Depending on the ultrasonic seam characteristics, it may be undesirable for the shadow (G) ores (AT), and consequently the weld depth, to exceed a certain level. In this case, an additional reference area parameter (for example, as) can be introduced, and a control line such as:

[0061] AT > as Poor seam can be added. Similarly, reductions can also be applied to the quality levels mentioned above.

[0062] Based on the reference range determined as a result of the comparison (for example, AT < as), the computer (5) processor provides a warning to stop the line and to detect and resolve the problem. In other cases, the system continues to operate.

[0063] In a different arrangement of the invention, if there are issues such as product slippage on the line, the camera (1 ) of the quality control system can be moved by a motor along the horizontal and vertical axes. In the event that the computer (5) processor of the system issues such a warning, the position or orientation of the camera (1 ) can be adjusted automatically or manually.

Claims

CLAIMS1 . An ultrasonic welding control system for controlling the seam quality of fabric subjected to ultrasonic welding by means of an image processing algorithm, comprising:• at least one camera (1) arranged to image the ultrasonic weld area (3) and to capture video images of the processed fabric,• at least one light source (2) positioned at a desired angle to create shadows (G) on the processed fabric,• a computer (5) configured to analyze the images of the shadowed fabric captured by the camera (1 ) using the embedded image processing algorithm, calculate the total shadow (G) area (AT), and determine the weld quality by comparing the calculated shadow (G) area (AT) with predefined reference area parameters (ai, 32, as, 34) stored in the system database to identify the weld quality level.

2. The ultrasonic welding control system according to claim 1 , wherein it comprises a light source (2) whose position and orientation angle are adjusted automatically or manually depending on the weld depth to be detected.

3. The ultrasonic welding control system according to claim 2, wherein the light source (2) is positioned on the line or mounted on a motorized or manual mechanical system for orientation adjustment.

4. The ultrasonic welding control system according to claim 1 , wherein it comprises a camera (1) that can be moved by a motor along the horizontal or vertical axis in the event of issues such as product slippage on the line.

5. The ultrasonic welding control system according to claim 4, wherein the camera (1 ) is positioned on the line or mounted on a motorized or manual mechanical system for orientation adjustment, in response to a warning issued by the computer (5) processor of the system in the event of issues such as product slippage on the line.

6. An alternative embodiment of the ultrasonic welding control system according to claim 1 , wherein it comprises a lightproof protective enclosure (4) in which thecamera (1 ) and light source (2) are positioned so that the shadow (G) detection part of the algorithm is not affected by ambient light intensity and color variations.

7. An alternative embodiment of the ultrasonic welding control system according to claim 1 , wherein it comprises a computer (5) performing shadow (G) detection using an artificial intelligence-based or advanced image processing-based algorithm, without the lightproof protective enclosure (4).

8. An ultrasonic welding control method for controlling the seam quality of fabric subjected to ultrasonic welding, via a system comprising at least one camera (1 ) imaging the ultrasonic weld area (3), at least one light source (2) positioned at an angle relative to the ultrasonic weld area (3), and a computer (5), comprising the steps of:• Projecting light at an angle onto the fabric subjected to ultrasonic welding to create shadows (G) on the fabric caused by depth,• Capturing video images of the shadowed fabric,• Creating image frames from the recorded videos,• Converting the format of the image frame for processing,• Blurring the image frame whose format has been changed,• Masking the image within a desired two-color range to enhance the visibility of the shadows (G) in the blurred image,• Defining the region to be analyzed for the masked shadows (G),• Detecting and marking the boundaries of the shadows (G) within the defined region,• Calculating the area of the marked shadows (G),• Comparing the calculated total shadow (G) area (AT) with predefined seam quality reference area parameters (81, 82, 83, 84) stored in the system database and determining the seam quality.

9. The ultrasonic welding control method according to claim 8, wherein if the calculated total shadow (G) area (AT) falls outside the determined reference area parameter values (ai, 82, as, 84) as a result of the comparison, a warning is issued to stop the line, detect the problem, and resolve it, or the system automatically stops itself.

10. The ultrasonic welding control method according to claim 8, wherein the seam quality is determined based on comparing the calculated total shadow (G) area (AT) with the predetermined reference area parameters (ai, 32, as, 34) as follows:

11. The ultrasonic welding control method according to claim 8, wherein the reference area parameters (ai, 32, as, 34) that determine the weld quality levels can be reduced or increased according to preference or need.

Images