Image processing-based and computer-aided quality control station

WO2024248743A1PCT designated stage expired Publication Date: 2024-12-05PRAGMATECH BİLGİ TEKNOLOJİLERİ SANAYİ & TİCARET ANONİM ŞİRKETİ
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
PCT/TR2023/050480
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Filing Date
2023-05-29
Publication Date
2024-12-05

Smart Images

  • Figure TR2023050480_05122024_PF_FP_ABST
    Figure TR2023050480_05122024_PF_FP_ABST
Patent Text Reader

Abstract

The invention is a computer-aided image processing-based quality control station that can perform color, size, and surface roughness check during production and at every stage of production with high precision, and can transfer the results to other applications such as stock tracking, characterized in that it is related to an isolated electronic unit, which has PLC and image processing technologies; an image acquisition unit, which transmits measurements together with large-diameter telecentric lenses; a servo precision adjustment unit, which enables the automatic centering of the product for the most precise accuracy by using servo motion during diameter measurements; a product identifier, which identifies the products in real time and transfers the information to the electronic unit; a centering section, which provides precise positioning of the product with motor-controlled centering from four points, just below the center point of the image acquisition unit; an operator panel, which has a touch screen and is used to export measurement data with a an easy-to-use interface, to control the operation of the device and to introduce new product models; and a multiple belt assembly which ensures the removal of the NOK (not okay) products from OK products and the separation of the products from each other.
Need to check novelty before this filing date? Find Prior Art

Description

[0001] IMAGE PROCESSING-BASED AND COMPUTER-AIDED QUALITY CONTROL STATION

[0002] Technical Field

[0003] The invention relates to a computer-aided image processing-based quality control station that can be applied to different industrial production fields, can automatically perform color, size, and surface roughness controls during production and at every stage of production with high precision, and can transfer the results to other applications of an enterprise such as resource management and stock tracking.

[0004] Prior art

[0005] In today's industry, production is carried out by following a number of process steps for the manufacture of a product. Each production batch is checked after production to ensure the quality of the resulting products. After performing the quality control tests on the samples taken to represent the production batches, the entire production batch containing the samples incapable of meeting the required criteria is deemed defective and the entire batch is discarded. This causes a tremendous loss of production, time, energy, and labor. This brings additional poor-quality costs to enterprises.

[0006] In a search conducted in the Turkish patent database, a number of relevant applications have been found:

[0007] - The abstract of an application numbered 2019 / 11772 indicated: The invention relates to a vehicle body quality control system that performs quality control processes such as part presence-absence, measurement control, etc. by means of a 2D comparison technique of a real image of large vehicles with an image in a virtual environment automatically and makes a decision accordingly. The abstract of the application numbered 2019 / 01395 indicates: "The invention relates to a continuous system and method that enables the transporting of the fabric, which has been printed in the rotating screen printing machine (10) in the textile industry to the fixing machine (30) by means of movable arms (20); the conveying of the same to the quality control automation unit (40) on the conveyor belt (70); the detection of pattern recognition and printing errors by means of the camera (42) connected with the image processing software (43) and the printing apparatus (44); the conveying of the same to the packaging process with the conveyor belt (70), thereby the separation of the defective products by means of the sorting and separation apparatus (51) and thusly the implementation of the packaging procedure."

[0008] The disadvantages of quality control methods used in today's industry are as follows:

[0009] • Since the samples are taken from the finished production batches, the discarded products go through all the stages of production, which creates a high cost of poor quality.

[0010] • In case of the inability of the representative sample from the production batch to meet the quality standards, all the products in the relevant production batch are deemed defective and discarded, and the quality ones are also wasted. This increases the cost of poor quality.

[0011] • Dimension and surface control with contact probes prolong the measurement. In addition, there are probe-related errors and challenges in taking measurements from complex-shaped parts in probed measurement.

[0012] • The visual control of color depends on many factors such as the personal, physical, and biological characteristics of the quality controller as well as the illumination of the measurement site and the point of view with a high margin of error.

[0013] • Carrying out size and surface control and color control at separate stations prolongs the control time and requires extra effort and time.

[0014] • Since the parts discarded as a result of the quality control are handled after production, depending on the recycling and waste management policies of enterprises such as the end of the day, weekend, or month-end, it is not possible to calculate the cost of poor quality instantly. In addition, stock management and waste management cannot be done instantly.

[0015] Description and Objects of the Invention

[0016] The invention is an improvement in the computer-aided image processing-based quality control station that can automatically perform color, size, and surface roughness controls during production and at every stage of production with high precision, and can transfer the results to other applications of an enterprise such as resource management and stock tracking. Furthermore, thanks to the barcode reading, QR code reading, and RFID tag reading features of the station, the product can be tracked simultaneously during production. As such, the invention has the potential to be the most critical and useful product for enterprises in the industry 4.0 process.

[0017] The invention first identifies the product by reading the barcode on the product arriving at the station, then measures the images obtained with two or more cameras in the system with high precision, makes color control, and can rapidly decide whether the product is defective by comparing the product with the quality requirements. The invention is supported by algorithms such as Pattern Matching, Geometric Model Finding, Shape Finding, Feature Extraction, Classification, Multi-dimensional Measurements, Color Analysis, Surface Continuity Analysis, Character / Text Recognition, Barcode and QR Code Extraction, and 3D Point Cloud Generation to implement this function.

[0018] The main object of the invention is to present a different design, which brings a new initiative in this field, unlike the methods used in the prior art.

[0019] Another aim of the invention is to provide an improved production system that detects poor-quality products in the early stages of production and prevents such products from passing to other production stages.

[0020] Another object of the invention is to ensure contact-free size, surface, and color control with high-precision image processing. If poor-quality products are detected as a consequence of the control, these products are discarded and sent for recycling, without being sent to the next stage of production.

[0021] Thus, since the products with size and surface defects are discarded before being sent to the painting process, unnecessary paint usage, energy waste, and cost are eliminated and the poor-quality cost is reduced.

[0022] Another object of the invention is to prevent the unnecessary discarding of high-quality products by controlling each product, not the whole batch in production. Since the station control time is a short time like 4 seconds, only defective products will be discarded, and quality parts in the production batch will continue to be processed on the production line. Thus, a high amount of scrap will be reduced.

[0023] Another object of the invention is to give a QR code to each product that has undergone quality control, to identify the product and introduce the same in the system, and to access all production and quality control information of the product in the following period. These data will be used in quality improvement processes in the future. In addition, the amount of discarded products is instantly output to the enterprise's software such as enterprise resource planning (ERP) and / or stock management, allowing instant stock and scrap follow-up.

[0024] Another object of the invention is to collect size and surface control and color controls in a single station, thusly performing the same at high speed and with high precision.

[0025] Another object of the invention is to completely eliminate the errors that may be due to the person due to the elimination of the human factor in quality control. In an example, the diameter of an object could be measured with an accuracy of + / - 10 microns, thanks to the 4-point centering system at the station.

[0026] Another object of the invention is to minimize the visual errors caused by lighting by using circular and linear illuminators. Thusly, a smoother image will be transferred to the cameras and possible errors will be prevented.

[0027] Drawings for Understanding the Invention

[0028] Fig. 1 Perspective view of the quality control station of the invention

[0029] Fig. 2: Detailed view of the components of the station

[0030] Part Numerals

[0031] 1 - Quality Control Station

[0032] 10- Image Acquisition Unit

[0033] 20- Electronic Unit

[0034] 30- Measurement Screen 40- Servo Precision Adjustment Unit

[0035] 50- Product Identifier

[0036] 60- Centering Section

[0037] 70- Operator Panel

[0038] 80- Multiple Belt 90- Profile Cabin

[0039] Detailed Description of the Invention

[0040] The invention is a computer-aided image processing-based quality control station (1) that can be applied to different industrial production fields, can automatically perform measurement, color, size, and surface roughness check during production and at every stage of production with high precision, and can transfer the results to other applications of an enterprise such as resource management and stock tracking, characterized in that it comprises an electronic unit (20), which has PLC and image processing technologies and is isolated to make it possible to use the same even in challenging environments; an image acquisition unit (10), which transmits perspective- free measurements together with large-diameter telecentric lenses; a servo precision adjustment unit (40), which enables the automatic centering of the product for the most precise accuracy by using servo motion during diameter measurements; a product identifier (50), which identifies the products in real time and transfers the information to the electronic unit (20); a centering section (60), which provides precise positioning of the product with motor-controlled centering from four points, just below the center point of the image acquisition unit (10); an operator panel (70), which has a touch screen and is used to export measurement data with a an easy-to-use interface, to control the operation of the device and to introduce new product models; and a multiple belt assembly (80) which ensures the removal of the NOK (not okay) products from OK products and the separation of the products from each other.

[0041] Fig. 1 details the components of the quality control station (1) of the invention. In the figure, the profile cabin (90), which is relatively easy to install and supported by durable Sigma-based profiles for long-term use, surrounds all the elements. There is an image acquisition unit (10) in the center of said profile cabin (90). This unit (10), together with large-diameter telecentric lenses, transfers the perspective-free measurements to the electronic unit (20). There is a centering section (60) under the image acquisition unit (10). The centering section (60) enables precise positioning of the product with motor- controlled centering from four points, just below the center point of the image acquisition unit (10). There is a measurement screen (30) for everyone to see the last measurement on the right side of the figure. There is an electronic unit (20) at the bottom of the figure. The electronic unit (20) is an isolated unit with PLC and image processing technologies in IP-X standard, enabling it to be used even in challenging environments.

[0042] There is a servo precision adjustment unit (40) positioned on the upper side of the image acquisition unit (10). This unit (40) enables the automatic centering of the product for the most precise accuracy by using servo motion during diameter measurements.

[0043] There is a product identifier (50) on the left side of said measurement screen (30). The product identifier (50) identifies the products in real-time with 99.99% accuracy and transfers this information to the electronic unit (20).

[0044] Fig. 2 shows a detailed view of the components of the quality control station (1 ). The detail of the measurement screen (30) is present at the top of the figure. There is an operator panel (70) behind the measurement screen (30).

[0045] The operator panel (70) has a touch screen and is used to export measurement data with an easy-to-use interface, control the operation of the device, and introduce new product models.

[0046] The is a multiple-belt (80) assembly at the bottom of the figure. The multiple-belt (80) ensures that NOK (not okay) products are removed from OK products so that the products do not mix with each other.

[0047] The components of an industrial example of the invention (1) and their positions in the figure have been mentioned above. The invention (1) is also a quality control method, comprising the process steps of:

[0048] Taking the product inside if the machine is empty, Determining the product model (50) before the product reaches the measurement area and informing the PLC unit by the PC that the product will be measured in the next stage, o Sending a message "Model could not be determined" to the PLC unit (20) if the product model could not be determined before the product reaches the measurement area, o Removing the product from the device without waiting in the measurement area if no other product is waiting at the device outlet by the PLC unit (20), and not taking any other product into the measurement area until the product leaves the machine, o For a model introduction (50) to the station (1 ):

[0049] ■ When the PLC unit (20) is working in the model identification (50) mode, taking a single product into the device (1 ) and positioning the product under the identification camera (50),

[0050] ■ Informing the PC that the first stage of the identification mode is ready and taking the necessary images for identification,

[0051] ■ Sending a message to the PLC to proceed to the next stage of the model identification by the PC,

[0052] ■ Taking the product to the measurement area by the PLC and conveying the message to the PC that the second stage of the identification mode is ready,

[0053] ■ After determining the necessary measurement values on the PC side, sending a message to the PLC indicating that the identification mode is finished,

[0054] ■ Removing the product from the device (1) by the PLC unit (20),

[0055] - Conveying to the PC that the product has entered the measurement area and is ready for measurement by the PLC unit (20), o If the product is OK:

[0056] ■ Providing information to the PLC unit that the product can proceed after the measurement on the PC side, ■ Removing the measured product from the machine if no other product is waiting at the device outlet by the PLC unit and not taking any other product into the measurement area until the product leaves the machine, e product is NOK:

[0057] ■ After the measurement made on the PC side, informing the PLC side that the product should be allocated to the REJECT belt,

[0058] ■ Separating the measured product into the REJECT belt by the PLC unit,

[0059] • If the PLC unit decides that the REJECT belt is full, stopping the line and taking no further action until the REJECT belt is emptied by the PLC unit,

[0060] • As soon as the REJECT belt is emptied, transferring the product to the REJECT belt and continuing the normal operation by the PLC unit.

Claims

CLAIMS1. A computer-aided image processing-based quality control station (1 ) that can be applied to different industrial production fields, can automatically perform measurement, color, size, and surface roughness controls during production and at every stage of production with high precision, and can transfer the results to other applications of an enterprise such as resource management and stock tracking, characterized in that it comprises:- an electronic unit (20), which has PLC and image processing technologies and is isolated to make it possible to use the same even in challenging environments;- an image acquisition unit (10), which transmits perspective-free measurements together with large-diameter telecentric lenses;- a servo precision adjustment unit (40), which enables the automatic centering of the product for the most precise accuracy by using servo motion during diameter measurements;- a product identifier (50), which identifies the products in real-time and transfers the information to the electronic unit (20);- a centering section (60), which provides precise positioning of the product with motor-controlled centering from four points, just below the center point of the image acquisition unit (10);- an operator panel (70), which has a touch screen and is used to export measurement data with an easy-to-use interface, control the operation of the device, and introduce new product models; a multiple belt assembly (80) which ensures the removal of the NOK (not okay) products from OK products and the separation of the products from each other.

2. The quality control station (1) according to claim 1 , characterized in that it comprises a profile cabin (90), which is relatively easy to install and supported by durable profiles for long-term use.

3. The quality control station (1) according to claim 1 , characterized in that it comprises a measurement screen (30) reflecting quality control measurement reports of the product.

4. The invention relates to a computer-aided image processing-based quality control method that can be applied to different industrial production fields, can automatically perform measurement, color, size, and surface roughness controls during production and at every stage of production with high precision, and can transfer the results to other applications of an enterprise such as resource management and stock tracking, comprising the process steps of:- Taking the product inside if the machine is empty,- Determining the product model before the product reaches the measurement area and informing the PLC unit by the PC that the product will be measured in the next stage,- Taking the product inside if the machine is empty,- Determining the product model (50) before the product reaches the measurement area and transferring the information to the PLC unit, the product's next measurement stage via the PC, o Sending a message "Model could not be determined" to the PLC unit (20) if the product model could not be determined before the product reaches the measurement area, o Removing the product from the device without waiting in the measurement area if no other product is waiting at the device outlet by the PLC unit (20), and not taking any other product into the measurement area until the product leaves the machine, o For a model introduction (50) to the station (1 ):■ When the PLC unit (20) is working in the model identification (50) mode, taking a single product into the device (1) and positioning the product under the identification camera (50),■ Informing the PC that the first stage of the identification mode is ready and taking the necessary images for identification,■ Sending a message to the PLC to proceed to the next stage of the model identification by the PC,■ Taking the product to the measurement area by the PLC and conveying the message to the PC that the second stage of the identification mode is ready,■ After determining the necessary measurement values on the PC side, sending a message to the PLC indicating that the identification mode is finished,■ Removing the product from the device (1) by the PLC unit (20),- Conveying to the PC that the product has entered the measurement area and is ready for measurement by the PLC unit (20), o If the product is OK:■ Providing information to the PLC unit that the product can proceed after the measurement on the PC side,■ Removing the measured product from the machine if no other product is waiting at the device outlet by the PLC unit and not taking any other product into the measurement area until the product leaves the machine, o If the product is NOK:■ After the measurement made on the PC side, informing the PLC side that the product should be allocated to the REJECT belt,■ Separating the measured product into the REJECT belt by the PLC unit,• If the PLC unit decides that the REJECT belt is full, stopping the line and taking no further action until the REJECT belt is emptied by the PLC unit, o As soon as the REJECT belt is emptied, transferring the product to the REJECT belt and continuing the normal operation by the PLC unit, Sending a message "Model could not be determined" to the PLC unit ifthe product model could not be determined before the product reaches the measurement area, o Removing the product from the device without waiting in the measurement area if no other product is waiting at the device outlet by the PLC unit, and not taking any other product into the measurement area until the product leaves the machine, o For a model introduction to the station:■ When the PLC unit is working in the model identification mode, taking a single product into the device and positioning the product under the identification camera,■ Informing the PC that the first stage of the identification mode is ready and taking the necessary images for identification,■ Sending a message to the PLC to proceed to the next stage of the model identification by the PC,■ Taking the product to the measurement area by the PLC and conveying the message to the PC that the second stage of the identification mode is ready,■ After determining the necessary measurement values on the PC side, sending a message to the PLC indicating that the identification mode is finished,■ Removing the product from the device by the PLC unit,- Conveying to the PC that the product has entered the measurement area and is ready for measurement by the PLC unit, o If the product is OK:■ Providing information to the PLC unit that the product can proceed after the measurement on the PC side,■ Removing the measured product from the machine if no other product is waiting at the device outlet by the PLC unit and not taking any other product into the measurement area until the product leaves the machine,e product is NOK:■ After the measurement made on the PC side, informing the PLC side that the product should be allocated to the REJECT belt, ■ Separating the measured product into the REJECT belt by thePLC unit,• If the PLC unit decides that the REJECT belt is full, stopping the line and taking no further action until the REJECT belt is emptied by the PLC unit,• As soon as the REJECT belt is emptied, transferring the product to the REJECT belt and continuing the normal operation by the PLC unit.

Citation Information

Patent Citations

  • Inspection system

    US20170008036A1

  • System and method for performing production line product identification

    US20180314867A1

  • Vision comparison inspection system

    US6434264B1