Ai-powered camera-based quality control system for continuous panel line

An AI-supported camera-based system addresses human error in sandwich panel production by continuously monitoring and identifying defects, ensuring reliable and efficient defect detection and analysis.

WO2025144224A1PCT designated stage Publication Date: 2025-07-03ASSAN PANEL SANAYI VE TICARET ANONIM SIRKETI
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
PCT/TR2024/050868
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Filing Date
2024-07-23
Publication Date
2025-07-03

AI Technical Summary

Technical Problem

Existing sandwich panel production systems rely on human visual inspection, which is prone to errors due to human factors such as shift changes and carelessness, leading to inconsistent defect detection.

Method used

An AI-supported camera-based quality control system with a high-resolution camera, lighting, and air cleaning mechanism, coupled with a database and AI software, continuously monitors panel surfaces, identifying defects and providing warnings.

Benefits of technology

Ensures consistent and automated defect detection, reducing human error and enabling retrospective analysis, thus improving the reliability and efficiency of defect identification.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure TR2024050868_03072025_PF_FP_ABST
    Figure TR2024050868_03072025_PF_FP_ABST
Patent Text Reader

Abstract

The invention relates to an artificial intelligence-supported camera-based quality control system that detects errors and defects on panel surfaces in the sandwich panel continuous production line and warns the personnel in charge, which comprises the following; camera (1), which progresses on the continuous panel line (10) and records the images of the sheet metal (8), which is the product whose defects will be detected, and transfers them to the computer (10), portal (6), which has a database where the image data recorded by the camera (1) is stored and where continuous traceability of these data and analysis data obtained from these data is ensured, artificial intelligence software (7), which identifies situations that can be described as defects on the sheet metal (8) and compares the images recorded by the camera (1) with the images related to the defect situations, computer (10), which is the element where the portal (6) and artificial intelligence software (7) are installed, the data is processed and displayed, and on which an audio or visual warning is given in case of an defect is detected on the sheet (8).
Need to check novelty before this filing date? Find Prior Art

Description

[0001] Al-Powered Camera-Based Quality Control System For Continuous Panel Line

[0002] Technical Field

[0003] The invention relates to an artificial intelligence-supported camera-based quality control system that detects faults and defects on panel surfaces in a sandwich panel continuous production line and prevents the continuation of the fault by alerting the personnel in charge.

[0004] Prior Art

[0005] In the prior art, in order to catch surface defects in sandwich panel production, employees are placed at certain stations and visual detection is made to catch the defects. However, in this application, it is not always possible to catch product defects when the employee leaves the observation point, there are no employees at these stations in cases such as shift changes, carelessness and human error margins are considered.

[0006] As a result, due to the abovementioned disadvantages and the insufficiency of the current solutions regarding the subject matter, a development is required to be made in the relevant technical field.

[0007] Purpose of the Invention

[0008] The invention aims to solve the abovementioned disadvantages by being inspired from the current conditions.

[0009] The main purpose of the invention is to introduce a quality control system that enables the detection of defects on the sheet metal, which will be accepted as errors in the process, by comparing the data obtained from the high-resolution camera, which is positioned at certain points and has image, light and motion stability, with artificial intelligence technology in a virtual environment and providing the relevant warnings.

[0010] A further purpose of the invention is to enable retrospective analyses to be performed by means of uninterrupted storage of the data obtained from the camera. Since the artificial intelligence software can also learn from errors, different defects that may arise in the future can be caught to a great extent thanks to the inventive system without a new definition. In order to fulfil the abovementioned aims, an artificial intelligence-supported camerabased quality control system that detects defects and defects on panel surfaces and warns the personnel in charge in the sandwich panel continuous production line has been developed, comprising of the following;

[0011] - a camera which moves along a continuous panel line, capturing images of the sheet metal and transferring them to a computer for defect detection,

[0012] - portal that has a database where the image data recorded by the camera is stored and where continuous traceability of these data and the analysis data obtained from these data is ensured,

[0013] - artificial intelligence software that identifies situations that can be characterised as errors on the sheet and compares the images recorded by the camera with the images of the error situations,

[0014] - computer on which the portal and artificial intelligence software is loaded, the data is processed and displayed, and an auditory or visual warning is given in case of errors detected on the sheet metal.

[0015] In a preferred embodiment of the invention, the quality control system comprises at least one lighting element which illuminates the environment to enable defects on the sheet metal to be recorded in a bright environment.

[0016] In a preferred embodiment of the invention, the quality control system comprises lightinsulating housing elements for preventing reflections of light from the lighting element on the sheet metal.

[0017] In a preferred embodiment of the invention, the quality control system comprises a camera stabiliser, which is the element to which the camera is connected, and which is fixed to the continuous panel line to ensure that the camera is stationary and in the desired position.

[0018] In a preferred embodiment of the invention, the quality control system comprises an air blowing tube for blowing air to remove dust from the lens of the camera.

[0019] The structural and characteristic features of the present invention will be understood clearly by the following drawings and the detailed description made with reference to these drawings and therefore the evaluation shall be made by taking these figures and the detailed description into consideration. Figures to Help Understanding of the Invention

[0020] Figure 1 is a two-dimensional side view of the system which is the subject of the invention.

[0021] Figure 2 is a perspective view of the system which is the subject of the invention.

[0022] Figure 3 is another perspective view of the system which is the subject of the invention.

[0023] Figure 4 is a two-dimensional top view of the system which is the subject of the invention.

[0024] Figure 5 is a perspective view of Figure 4.

[0025] Description of the Part References

[0026] 1 . Camera

[0027] 2. Air blowing pipe

[0028] 3. Housing element

[0029] 4. Lighting element

[0030] 5. Camera stabilizer

[0031] 6. Portal

[0032] 7. Artificial intelligence software

[0033] 8. Sheet metal

[0034] 9. Continuous panel line

[0035] 10. Computer

[0036] Detailed Description of the Invention

[0037] In this detailed description, the preferred embodiments of the inventive artificial intelligence supported camera-based quality control system on continuous panel line (9) are described by means of examples only for clarifying the subject matter.

[0038] The preferred embodiment of the system of the present invention comprises the following elements:

[0039] - camera (1 ),

[0040] - air blowing pipe (2)

[0041] - housing element (3),

[0042] - lighting element (4), - camera stabilizer (5),

[0043] - portal (6),

[0044] - artificial intelligence software (7),

[0045] - sheet metal (8),

[0046] - computer (10).

[0047] The features of the elements of the system that is the subject of the invention are given below:

[0048] - The camera (1 ) is preferably positioned under the route where the sheet metal moves in the continuous panel line (9), and can record the sheet metal (8), which is the product whose defects will be detected. As sheet metal moves on the continuous panel line (9), the camera (1 ) records these defects in order to transfer the defects on the sheet metal to the computer (10) wired or wirelessly. The camera is preferably high resolution.

[0049] - The air blowing pipe (2) blows air onto the lens of the camera (1 ) and the upper region of the lens to clean the dust coming on the lens of the camera (1)-

[0050] - The housing element (3) provides light insulation and prevents reflections on the sheet metal (8) caused by the light coming from the lighting elements (4). The lighting element (4) is the element which illuminates the environment in order to enable the defects on the sheet metal (8) to be visualised in a bright environment.

[0051] - The camera stabiliser (5) is the element to which the camera (1 ) is connected in order to ensure that the camera (1 ) is stationary and in the desired position, and the camera stabiliser (5) is continuously fixed to the panel line.

[0052] - The portal (6) is a platform that has a database where the image data recorded by the camera are stored and continuous monitoring of these data and the analysis data obtained from these data is ensured.

[0053] - The artificial intelligence software (7) is the software that identifies the situations that can be characterised as errors on the sheet (8) and compares the images recorded by the camera (1 ) with the images of the error situations.

[0054] - The computer (10) is the element where the portal (6) and the Al software (7) are loaded, and then the data is processed and displayed. If an error is detected on the sheet (8), an audible or visual warning is given via the computer (10). The working principle of the quality control system subject to the invention is as follows: The camera (1) records the sheet metal (9) travelling along the continuous panel line (10). The recorded images are transmitted to the computer (10). These images transferred to the computer (10) are compared by the artificial intelligence software (7) with the error conditions determined in the artificial intelligence software (7), and if an error is detected, an auditory or visual warning is given via the computer (10). Thus, it is ensured that the personnel on duty are warned.

[0055] During image recording of the camera (1), in order to keep the lens of the camera (1) clean, the air blowing tube (2) blows air into the lens of the camera (1 ) and the upper area of the lens to remove dust on the lens of the camera (1).

[0056] On the other hand, at least one lighting element (4) illuminates the lower region of the sheet metal (8) and the region where the camera (1 ) is located, so that the defects on the sheet metal (8) are recorded by the camera (1) in a bright environment. In order to be able to fully visualise the defects on the sheet metal (8), a closed environment is created by means of housing elements (3), preferably in the form of plates, preferably positioned on both sides of the camera (1 ) and at the bottom of the sheet metal (8), in order to ensure good illumination of the environment by the lighting element (4). Thus, the housing elements (3) prevent the reflection of light from the lighting elements (4) on the sheet metal (8).

Claims

CLAIMS1. Artificial intelligence supported camera-based quality control system that detects faults and defects on panel surfaces and warns the personnel in charge in the sandwich panel continuous production line, characterized by comprising;- a camera (1 ) which records the images of the sheet (8), which is the product to be detected defects and which advances in the continuous panel line (10), and transfers them to the computer (10),- a portal (6) that has a database where the image data recorded by the camera (1 ) is stored and where continuous traceability of these data and the analysis data obtained from these data is ensured,- artificial intelligence software (7) that identifies situations that can be characterised as errors on the sheet metal (8) and compares the images recorded by the camera (1 ) with the images of the error situations,- computer (10) on which the portal (6) and artificial intelligence software (7) is loaded, the data is processed and displayed, and an auditory or visual warning is given in case of errors detected on the sheet metal (8).

2. The quality control system according to claim 1 , characterized by comprising at least one lighting element (4) illuminating the environment in order to ensure that defects on the sheet metal (8) are recorded in a bright environment.

3. The quality control system according to claim 2, characterized by comprising lightinsulating housing elements (3) for preventing reflections of light from the lighting element (4) on the sheet metal (8).

4. The quality control system according to any one of the preceding claims, characterized by comprising a camera stabiliser (5), which is the element to which the camera (1 ) is connected, and which is fixed to the continuous panel line in order to ensure that the camera (1 ) is fixed and in the desired position.

5. The quality control system according to any one of the preceding claims, characterized by comprising an air blowing tube (2) for blowing air to remove dust from the lens of the camera (1 ).

Citation Information

Patent Citations

  • Visual inspection and identification equipment for stokehole insert

    CN209069836U

  • Circuit board terminal detection equipment

    CN218766565U

  • Method and apparatus for component fault detection based on image

    US20210073973A1