KR20230078412A - AI-assisted Reliability Assessment Device for Printing, Reliability Assessment Method using the same, and the Printing System - Google Patents

Abstract

The present invention relates to an AI-based printing reliability evaluation device, a reliability evaluation method using the same, and a printing system capable of evaluating reliability.
To this end, the present invention is characterized in that printing reliability is evaluated using convolutional neural network models (CNNs) that receive a print pattern image of a printing device to be evaluated and learn whether or not the print pattern is defective.
Accordingly, accurate and precise evaluation of print reliability is implemented, and reliability of the print pattern may be increased by predicting the possibility of defects of the print pattern in advance.

Description

Al-assisted Reliability Assessment Device for Printing, Reliability Assessment Method using the same, and the Printing System}

The present invention relates to an AI-based printing reliability evaluation device, a reliability evaluation method using the same, and a printing system capable of evaluating reliability, and more particularly, AI for predicting printing defects that may occur due to repeated substrate printing in advance. It relates to a printing reliability evaluation device based on the same, a reliability evaluation method using the same, and a printing system capable of evaluating reliability.

In the substrate printing of electronic devices, it is most important to form a homogeneous printed pattern. Since defects in printing can degrade the performance of the electronic device and cause a failure, the printed pattern must be evaluated in real time to confirm that there are no defects in the printing device.

On the other hand, the gravure offset printing device is a printing technology used in various electronic devices such as grid transparent electrodes, pressure sensors, and flat inductors, and can be implemented with a resolution as fine as the unit of printing reaches a micrometer, and electroluminescence It is widely used to print non-planar electronic devices such as displays (ELDs) and radio frequency identification (RFID) antennas.

This gravure offset printing proceeds in the order shown in (a) to (h) of FIG. 1, first filling the gravure plate (or roller) with ink (see (a), (b)) and ink doctoring After performing (see (c)), ink is transferred from the gravure plate using a blanket roller made of silicon (off process, see (d) to (f)). Thereafter, the ink transferred to the blanket roller is transferred to the substrate to form a plurality of print lines on the substrate (set process, see (g) to (h)).

Figure 2 shows the ink delivery mechanism in the (a) off process and (b) set process in gravure offset printing to make it easier to understand. do. Therefore, in order to form an ideal print pattern, 100% of the ink transferred to the blanket roller should be able to be transferred to the substrate, and this can be guaranteed when the cohesive force of the ink itself and the adhesive force to the substrate are higher than the adhesive force of the blanket roller and the ink.

However, in the printing process, the solvent of the ink is absorbed into the silicone material of the blanket roller, but the solvent absorption reaches saturation through repeated printing and is no longer absorbed, thereby reducing the cohesion of the ink itself. As a result, a phenomenon occurs in which ink is not transferred to the substrate in a state in which ink remains on the blanket roller. These printing defects affect the next set process in the repeated offset process, resulting in expansion of the width of the printed line, formation of protrusions, poor surface roughness, and the like. As an example, (a) of FIG. 3 is a printed line without defects, and (b) is a printed line with defects.

On the other hand, conventionally, a printed pattern formed on a substrate is monitored in real time to detect a printed pattern that is out of tolerance based on the width. However, since the conventional print pattern detection method completely relies only on the width of the printed line, there is a limitation in that some defects cannot be detected by the conventional image processing device.

Republic of Korea Patent Publication No. 10-2009-0089070 (published on March 29, 2011) Republic of Korea Patent Publication No. 10-2010-0048921 (published on 2011.12.02) Republic of Korea Patent Publication No. 10-2010-0072327 (published on December 14, 2010) Republic of Korea Patent Publication No. 10-2010-0126201 (published on December 1, 2010) Republic of Korea Patent Publication No. 10-2010-0126918 (published on May 29, 2012) Republic of Korea Patent Publication No. 10-2011-0067433 (published on December 9, 2011) Republic of Korea Patent Publication No. 10-2012-0016317 (published on August 27, 2013) Republic of Korea Patent Publication No. 10-2012-0144212 (published on June 20, 2014)

The present invention aims to solve the above problems and other problems related thereto.

An exemplary object of the present invention is to use an AI model that has previously learned whether or not a print pattern is defective so that a homogeneous print pattern can be formed by a printing device, thereby predicting the possibility of a print pattern defect in advance and printing It is possible to increase the reliability of the pattern, and to check the printing pattern in real time to control the driving of the printing device.

The technical tasks to be achieved by the AI-based printing reliability evaluation device according to the technical idea of the technology disclosed in this specification, the reliability evaluation method using the same, and the printing system capable of evaluating reliability are not limited to the above-mentioned tasks, and are not limited to those mentioned above. Other tasks will become apparent to those skilled in the art from the description below.

An AI-based printing reliability evaluation apparatus according to an embodiment of the present invention receives a printing pattern image of a printing device to be evaluated and uses convolutional neural network models (CNNs) to learn whether or not a printing pattern is defective. is configured to evaluate

Also, the printed pattern may be a printed pattern including a plurality of printed lines.

In addition, printing reliability can be evaluated by identifying a change in the width (W) of the printed line.

In addition, bulges may be identified based on the change in the width (W) of the printed line, and the number of identified bulges may be calculated to evaluate printing reliability.

An AI-based printing reliability evaluation device according to another embodiment of the present invention receives a printing pattern image of a gravure offset printing device to be evaluated, identifies a change in the width (W) of a printing line, and evaluates printing reliability is configured to

In addition, printing reliability may be evaluated by identifying bulges based on a change in the width (W) of the printed line and calculating the number of identified bulges.

In addition, printing reliability can be evaluated using convolutional neural network models (CNNs) that have learned whether or not a print pattern has defects.

Meanwhile, an AI-based printing reliability evaluation method according to an embodiment of the present invention includes the steps of receiving, by a printing reliability evaluation device, a print pattern image of a printing device to be evaluated; and evaluating print reliability using convolutional neural network models (CNNs) that have learned whether or not the print pattern is defective.

The evaluating of the printing reliability may include identifying a bulge based on a change in the width (W) of the printed line; and evaluating the printing reliability by calculating the number of identified bulges.

Meanwhile, a printing system capable of evaluating printing reliability according to an embodiment of the present invention includes a printing device; a control device for controlling the printing device and photographing a print pattern image; and an evaluation device that receives the print pattern image captured by the control device, evaluates printing reliability, and transmits the result to the control device.

Also, the control device may control driving of the printing device based on the printing reliability evaluated by the evaluation device.

In addition, the evaluation device may provide the evaluated printing reliability information in the form of information for producing a report.

In addition, the control device may generate the form of information for producing a report provided by the evaluation device in the form of a final report.

In addition, it may further include a network driver that is connected to the control device and the evaluation device by wire or wireless to store information.

In addition, the printing apparatus, the gravure plate; a doctoring unit filling the gravure plate with ink and performing doctoring; A blanket roller made of silicone; And a solvent removal unit; may include.

The solvent removing unit may remove a solvent of the ink absorbed by the blanket roller.

According to an embodiment of the present invention, the possibility of a print pattern defect is predicted in advance by evaluating the print reliability using a convolutional neural network model (CNNs) that receives a print pattern image of a printing device and learns whether or not the print pattern is defective. Thus, the reliability of the printed pattern can be increased.

In particular, there is an advantage in that efficient prediction is possible by evaluating the printing reliability by calculating the number of defects regardless of the defect location or size of the printed pattern.

Furthermore, by interlocking the printing device with the control device for controlling the printing device as well as the evaluation device, it is possible to check the printing pattern in real time and control driving of the printing device.

On the other hand, the effects described above are merely exemplary, and effects predicted or expected from the detailed configuration of the present invention from the viewpoint of those skilled in the art may also be added to the unique effects of the present invention.

1 is a conceptual diagram illustrating the operation principle of gravure offset printing in time series.
2 is a conceptual diagram illustrating an ink delivery mechanism made in gravure offset printing.
3 is an image of a printed pattern without defects and a printed pattern with defects according to an exemplary embodiment.
4 is a block diagram showing a printing system according to an embodiment of the present invention.
5 is a configuration diagram showing a printing apparatus according to an embodiment of the present invention.
6 is a block diagram illustrating a DNN model for image classification using a Skip Connections algorithm according to an embodiment.
7 is a graph illustrating a learning process of an image classification model according to an exemplary embodiment.
8 is an image showing a matrix for a verification data set according to an embodiment.
9 is original images ((a), (b)) and Grad-CAM images ((c), (d)) of printed patterns according to an embodiment.
10 is a graph illustrating clustering for YOLOv3 anchor boxes.
11 is a graph illustrating the YOLOv3 object detection model training process.
12 is a flowchart illustrating a sequence of fine-tuning and prediction of a YOLOv3 object detection model according to an embodiment of the present invention.
13 is an image in which printing defects are identified in a print pattern image according to an embodiment of the present invention.
14 is a graph showing the relationship between the width of a printed line and the number of defects.
15 is a flowchart illustrating a printing reliability evaluation method according to an embodiment of the present invention.

Hereinafter, the embodiments disclosed in this specification will be described in detail with reference to the accompanying drawings, but the same reference numerals will be assigned to the same or similar components regardless of reference numerals, and overlapping descriptions thereof will be omitted. The suffixes "module" and "unit" for components used in the following description are given or used together in consideration of ease of writing the specification, and do not have meanings or roles that are distinct from each other by themselves.

In addition, in describing the embodiments disclosed in this specification, if it is determined that a detailed description of a related known technology may obscure the gist of the embodiment disclosed in this specification, the detailed description thereof will be omitted. In addition, the accompanying drawings are only for easy understanding of the embodiments disclosed in this specification, the technical idea disclosed in this specification is not limited by the accompanying drawings, and all changes included in the spirit and technical scope of the present invention , it should be understood to include equivalents or substitutes.

It is understood that when an element is referred to as being "connected" or "connected" to another element, it may be directly connected or connected to the other element, but other elements may exist in the middle. It should be. On the other hand, when an element is referred to as “directly connected” or “directly connected” to another element, it should be understood that no other element exists in the middle.

Singular expressions include plural expressions unless the context clearly dictates otherwise.

In this application, terms such as "comprise" or "have" are intended to designate that there is a feature, number, step, operation, component, part, or combination thereof described in the specification, but one or more other features It should be understood that the presence or addition of numbers, steps, operations, components, parts, or combinations thereof is not precluded.

While the present disclosure has been illustrated and described in detail in the drawings and foregoing description, the present disclosure is to be considered illustrative rather than restrictive in nature, with only certain embodiments shown and described, and within the spirit of the present disclosure. It will be appreciated that all changes and modifications that enter are preferably protected.

Printing system capable of evaluating printing reliability

First, referring to FIGS. 4 and 5, an AI-based printing reliability evaluation apparatus according to an embodiment of the present invention, a printing reliability evaluation method using the same, and a printing system including the same capable of evaluating printing reliability will be described. do.

4 is a block diagram showing a printing system according to an embodiment of the present invention, and includes a printing device 100, a control device 200, and an evaluation device 300.

The printing device 100 is a device for forming a conductive printed pattern on a substrate, and may be one of various types of substrate printing devices. Preferably, the printing device 100 may be a gravure offset printing device in which a printed pattern to be formed has a plurality of printing lines.

5 is a configuration diagram showing a printing apparatus 100 according to an embodiment of the present invention, wherein the printing apparatus 100 includes a gravure plate 110, a doctor ring unit 120, a blanket roller 130, a stage (140).

The gravure plate 110 is manufactured in a plate shape so as to be relatively movable with the blanket roller 130 or manufactured in a rotating roller shape to accommodate ink. The doctoring unit 120 is doctoring so that the surface of the ink transferred from the gravure plate 110 is cleaned, and depending on the embodiment, the doctoring unit 120 itself accommodates the ink and gravure It may be manufactured to deliver ink to the plate 110 .

The blanket roller 130 is preferably made of silicon, receives ink from the gravure plate 110 and transfers it to the substrate S. The blanket roller 130 is rotated to transfer ink and can move up and down so that its relative position with the gravure plate 110 or the doctor ring unit 120 can be continuously changed.

The gravure plate 110, the doctor ring unit 120, the blanket roller 130, and the substrate S of the printing apparatus 100 according to an embodiment of the present invention are provided on the stage 140 so that printing can be performed. The gravure plate 110, the doctor ring part 120, and the blanket roller 130 are manufactured to realize relative positional movement by a driving means.

In addition, the control device 200 is for controlling the driving of the printing device 100, and controls the printing pressure and printing speed based on the load cell as well as turning on/off the driving of the printing device 100 described above. And, the relative position movement of the above-described gravure plate 110, doctor ring unit 120, blanket roller 130, and substrate S is implemented, and the ink is automatically filled by checking the amount of ink.

On the other hand, according to an embodiment of the present invention, as shown in the block diagram of FIG. 4, the control device 200 controls the printing device 100 and takes pictures of printed pattern images formed on the substrate S. module 210. The photographing module 210 preferably has a resolution of at least 2.41 μm/pixel so as to sufficiently identify the printed pattern.

In addition, the evaluation device 300 receives the print pattern image photographed by the control device 200, evaluates the printing reliability, and then transmits the evaluated printing reliability information to the control device 200 again. At this time, it is preferable that the evaluation device 300 evaluates the printing reliability using convolutional neural network models (CNNs) that receive the printed pattern image of the printing device 100 to be evaluated and learn whether or not the printing pattern is defective. do.

In addition, as shown in FIG. 4 , the printing system of the present invention may further include a network driver 400 connected to the control device 200 and the evaluation device 300 by wire or wireless to store information. The control device 200 may store the printed pattern image of the board captured by the photographing module 210 in the network driver 400, and the evaluation device 300 may store the printed pattern image stored in the network driver 400. The printing reliability information is retrieved, evaluated, and the evaluated printing reliability information is stored in the network driver 400 so that the control device 200 can call the printing reliability information stored in the network driver 400 again.

On the other hand, the evaluation device 300 may provide the evaluated print reliability information in the form of information for report production, and the information form for report production may include a csv file and an image file as shown in FIG. can In addition, the control device 200 can generate a final report in the form of a final report so that the manager can visually check it by using the form of information for report production provided by the evaluation device 300, and the final report can be generated by the network driver 400. ) can be stored in

Also, the control device 200 may control driving of the printing device 100 based on the printing reliability information evaluated by the evaluation device 300 . The control device 200 can turn on/off the driving itself of the printing device 100 based on the printing reliability information, and can appropriately control printing pressure and printing speed.

According to one embodiment, the printing apparatus 100 may additionally include a solvent removal unit 150 as shown in FIG. 5 . When the ink absorbed by repeated use of the blanket roller 130 reaches saturation and no longer absorbs the solvent, the solvent removal unit 150 removes the solvent of the absorbed ink, thereby improving self-cohesion of the ink. and secures the adhesion of the substrate (S) so that complete transfer can be achieved.

At this time, it is preferable that the solvent removal unit 150 is disposed adjacent to the blanket roller 130 to automatically remove the solvent absorbed by the blanket roller 130 by the control device 200, but it is manufactured as a separate device. It is also possible to operate the solvent removal unit 150 by a manager according to a signal from the controller 200 .

The printing system capable of evaluating printing reliability described above checks the printing pattern in real time by interlocking the printing device 100 with the control device 200 that controls the printing device as well as the evaluation device 300 to operate the printing device 100. Since it can be controlled, it is possible to provide a print pattern with uniform quality.

AI-based printing reliability evaluation device and evaluation method

Hereinafter, the above-described evaluation device 300 will be described in more detail. The AI-based printing reliability evaluation apparatus 300 according to an embodiment of the present invention is a convolutional neural network model (CNNs) that learns whether or not a printing pattern is defective by receiving a printing pattern image of the printing apparatus 100 to be evaluated. Evaluate printing reliability using

To this end, as shown in FIG. 3, the present invention classifies (a) a plurality of print pattern images without defects and (b) a plurality of print pattern images with defects in advance and assigns a label to each of them, thereby verifying the verification. Build a data set for Table 1 below is a table showing data sets built according to an embodiment, and distributions of image data sets for image classification and object detection tasks.

Meanwhile, FIG. 6 is a block diagram illustrating a DNN model for image classification using an algorithm of skip connections by adopting the largest number of layers. The built data set is augmented including arbitrary brightness, contrast, zoom, rotation enhancement layer, and arbitrary horizontal and vertical flipping for expansion, and weighting can be given considering the imbalance of classified data. .

7 shows a learning process of an image classification model, and a predictive model with excellent accuracy can be adopted and utilized by performing a plurality of epochs on an extended data set for learning. The predictive model was evaluated using a validation data set, and the results can be displayed as a matrix as shown in FIG. 8 .

You can also visualize the trained predictive model using the Grad-CAM approach to reassure model performance, and see which parts of the image influenced the outcome based on the last convolutional layer input related to the predictive model. can be verified (a) and (c) of FIG. 9 show the original image of the print pattern without defects and the Grad-CAM image, respectively, and (b) and (d) show the original image of the print pattern with defects and the Grad-CAM image. This is a CAM image, and the Grad-CAM image shown in (c) and (d) of FIG. 9 can be overlapped with the original image.

The above learned prediction model can be applied with a fine-tuning approach to retrain the YOLOv3 model. For example, as shown in FIG. 10, IoU (Intersection Over Union) between the center of each cluster and the data points can be calculated by analyzing and normalizing a bounding box in an N × N pixel image and performing clustering. can

11 shows the object detection model learning process of the YOLOv3 model. In the present invention, it is possible to store the best weights with the highest mAP and build the YOLOv3 model by performing thousands of repetitions of training using a data set. 12 shows a flow chart of YOLOv3 object detection model fine-tuning and prediction according to one embodiment of the present invention.

According to the printing reliability evaluation apparatus 300 of the present invention, printing defects can be predicted in a printing pattern image as shown in FIG. 13 using the convolutional neural network models (CNNs) described above. Meanwhile, the evaluation device 300 may evaluate printing reliability by identifying a change in the width W of the printed line, since the possibility of a defect appearing increases as the width W of the printed line increases.

However, the print reliability evaluation apparatus 300 of the present invention calculates the number of defects regardless of the location or size of defects formed on a printed line. 14 is a graph showing a result of comparing a change in the width (W) of a printed line detected by an AI model and a change in the number of defects. That is, the printing reliability evaluation apparatus 300 of the present invention may evaluate printing reliability by identifying bulges based on a change in the width (W) of the printing line and calculating the number of identified bulges. As a result, the number of defects is calculated regardless of the defect location or size of the printed pattern to evaluate the printing reliability, thereby providing an advantage of efficient prediction.

On the other hand, the printing reliability evaluation device 300 of the present invention described above can be applied to various types of substrate printing devices, but preferably, the printing device 100 is a gravure offset having a plurality of printing lines in the printing pattern to be formed ( Gravure Offset) may be a printing device.

In addition, in the AI-based printing reliability evaluation method (M) according to an embodiment of the present invention, as shown in FIG. It includes a step of receiving a print pattern image (S10) and a step of evaluating the printing reliability of the photographed print pattern image using convolutional neural network models (CNNs) that have learned whether or not the print pattern is defective (S20). .

In particular, the step of evaluating the printing reliability (S20) includes the step of identifying a bulge based on the change in the width (W) of the printed line (S21) and calculating the number of defined bulges to evaluate the printing reliability. It may proceed including the step (S22) of doing.

The above-described device and its control may be implemented as hardware components, software components, and/or a combination of hardware components and software components. For example, devices and components described in the embodiments include a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable gate array (FPGA), and a programmable PLU (programmable logic unit). logic unit), microprocessor, or any other device capable of executing and responding to instructions. The processing device may run an operating system (OS) and one or more software applications running on the operating system. A processing device may also access, store, manipulate, process, and generate data in response to execution of software. For convenience of understanding, there are cases in which one processing device is used, but a person skilled in the art will understand that the processing device may include a plurality of processing elements and/or a plurality of types of processing elements. it can be seen that there is For example, a processing device may include a plurality of processors or a processor and a controller. Other processing configurations are also possible, such as parallel processors.

Software may include a computer program, code, instructions, or a combination of one or more of the foregoing, which configures a processing device to operate as desired or, independently or in combination, instructs the processing device. can do. Software and/or data may be permanently or temporarily stored in any tangible machine, component, physical device, virtual device, computer storage medium or device for interpretation by a processing device or to provide instructions or data to a processing device. can materialize. Software may be distributed on networked computer systems and stored or executed in a distributed manner. Software and data may be stored on one or more computer readable media.

The method according to the embodiment may be implemented in the form of program instructions that can be executed through various computer means and recorded on a computer readable medium. The computer readable medium may include program instructions, data files, data structures, etc. alone or in combination. Program commands recorded on the medium may be specially designed and configured for the embodiment or may be known and usable to those skilled in computer software. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks and magnetic tapes, optical media such as CD-ROMs and DVDs, and magnetic media such as floptical disks. - includes hardware devices specially configured to store and execute program instructions, such as magneto-optical media, and ROM, RAM, flash memory, and the like. Examples of program instructions include high-level language codes that can be executed by a computer using an interpreter, as well as machine language codes such as those produced by a compiler. The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.

As described above, although the embodiments have been described with limited examples and drawings, those skilled in the art can make various modifications and variations from the above description. For example, the described techniques may be performed in an order different from the method described, and/or components of the described system, structure, device, circuit, etc. may be combined or combined in a different form than the method described, or other components may be used. Or even if it is replaced or substituted by equivalents, appropriate results can be achieved.

Therefore, other implementations, other embodiments, and equivalents of the claims are within the scope of the following claims.

100: printing device 110: gravure plate
120: doctor ring unit 130: blanket roller
140: stage 150: solvent removal unit
200: control device 210: shooting module
300: evaluation device S: substrate

Claims (16)

Characterized in that the printing reliability is evaluated using convolutional neural network models (CNNs) that receive the printing pattern image of the printing device to be evaluated and learn whether or not the printing pattern is defective,
AI-based printing reliability evaluation device.
According to claim 1,
Characterized in that the printed pattern is a printed pattern including a plurality of printed lines,
AI-based printing reliability evaluation device.
According to claim 2,
Characterized in that the print reliability is evaluated by identifying the change in the width (W) of the printed line,
AI-based printing reliability evaluation device.
According to claim 3,
Characterized in that a bulge is identified based on a change in the width (W) of the printed line, and the printing reliability is evaluated by calculating the number of identified bulges.
AI-based printing reliability evaluation device.
Characterized in that the print reliability is evaluated by receiving the print pattern image of the gravure offset printing device to be evaluated and calculating the change in the width (W) of the print line,
AI-based gravure offset printing reliability evaluation device.
According to claim 5,
Characterized in that a bulge is identified based on a change in the width (W) of the printed line, and the printing reliability is evaluated by calculating the number of identified bulges.
AI-based gravure offset printing reliability evaluation device.
According to claim 5,
Characterized in that the printing reliability is evaluated using convolutional neural network models (CNNs) that have learned whether or not the printing pattern is defective,
AI-based gravure offset printing reliability evaluation device.
A printing reliability evaluation device,
receiving a print pattern image of a printing device to be evaluated; and
Evaluating print reliability using convolutional neural network models (CNNs) that have learned whether or not the printed pattern is defective; characterized in that it includes,
AI-based printing reliability evaluation method.
According to claim 8,
The step of evaluating the printing reliability,
identifying a bulge based on a change in the width (W) of the printed line; and
Evaluating the printing reliability by calculating the number of defined bulges; characterized in that it includes,
AI-based printing reliability evaluation method.
printing device;
a control device for controlling the printing device and photographing a print pattern image; and
characterized in that it comprises; an evaluation device that receives the print pattern image taken by the control device, evaluates the printing reliability, and transmits it to the control device.
A printing system capable of evaluating printing reliability.
According to claim 10,
The control device,
Characterized in that the driving of the printing device is controlled based on the printing reliability evaluated by the evaluation device.
A printing system capable of evaluating printing reliability.
According to claim 10,
The evaluation device,
Characterized in that the evaluated printing reliability information is provided in the form of information for report production,
A printing system capable of evaluating printing reliability.
According to claim 12,
The control device,
Characterized in that the form of information for producing a report provided by the evaluation device is generated in the form of a final report,
A printing system capable of evaluating printing reliability.
According to claim 10,
Characterized in that it further comprises; a network driver connected to the control device and the evaluation device by wire or wireless to store information.
A printing system capable of evaluating printing reliability.
According to claim 10
The printing device,
gravure plate;
a doctoring unit filling the gravure plate with ink and performing doctoring;
a blanket roller made of silicone; and
Characterized in that it comprises a; solvent removal unit
A printing system capable of evaluating printing reliability.
According to claim 15
The solvent removal unit,
Characterized in that the solvent of the ink absorbed by the blanket roller is removed.
A printing system capable of evaluating printing reliability.

Images