KR100351129B1 - System for Examination printing inferior using Circumference Mask - Google Patents

Abstract

The present invention is a printing defect inspection system using a peripheral mask for applying a peripheral mask to the image of the printed object to be printed on the printed object to more reliably perform the defect inspection of the printed material adaptively to various changing printing conditions To provide.

To this end, the present invention in the inspection system for checking whether the printed matter printed on a predetermined printing object by the operation of the printing machine, print image input means for receiving the image of the printed image and the image of the printed image Image processing means for correcting and binarizing, model region setting means for setting an image of a standard model region for comparison with an image of a test subject among the printed images, and generating a peripheral mask image set by varying the size of the standard model image And calculating a difference image of the image of the standard model and the printed image of the inspection target, and combining the difference image and the peripheral mask image to determine whether the printing is defective by analyzing the occurrence of blobs. Visual inspection status of printed image with peripheral mask And display means for displaying the data, and data communication means for transmitting the test result data according to the print defect determination result by the control means to the printer.

Description

System for Examination printing inferior using Circumference Mask}

The present invention relates to a printing defect inspection system using a periphery mask, and more particularly, by applying a periphery mask to a printed image printed on a printed object of various materials, to offset the determination of fine errors due to the printing conditions of the printed matter. The present invention relates to a printing defect inspection system using a peripheral mask to enable a faster and more efficient determination of printing defects.

In general, in the manufacturing field of manufacturing and mass-producing various products by using various elastic and non-elastic materials, the brand's trademark, design or manufacturing mark of each product is printed in the form of printed material, thereby spreading the awareness of the product. In addition to planning, the production status and characteristics of goods can be grasped through printed materials.

The printing apparatus for printing a printed matter on a product as a print object is adopted a high speed printing machine which prints at a high speed and in a large quantity in order to meet the mass production method of the printed object. It is possible to determine the error state of the printed matter printed on the printing object through a separate print defect inspection device.

A conventional printing defect inspection apparatus for determining an error state of a printed matter converts image information obtained from a printed matter into a reasonable image, sets it as an image of a standard model, and inputs from the printed image printed on the image of the standard model and the printed object. It is possible to determine the error state by obtaining the difference between one image, the relation for the difference comparison is as shown in Equation 1 below.

Here, T represents an image (n × m) of the entire template, and I represents an input image (N × M) to be examined.

According to the difference comparison method of the printed image using the relational expression of Equation 1, as shown in FIG. 1A, an image of a standard model is set for a predetermined printed matter, and the image of the standard model is printed on a print object. The printing defect can be determined by comparing the difference with the actual printed image.

The detection of the differential image for comparing the image of the standard model and the image of the printed matter is to find a blob by simply comparing the image of the standard model and the printed image of the inspection target, as shown in Figs. 2a to 2c, If the blob is analyzed and its size is above a predetermined threshold, printing is judged to be defective, or the peripheral length and width of the blob are determined, and if the area is larger than the threshold length, the defect can be determined to be defective. .

Usually, the determination of a print defect on a printed matter uses the difference of the image as described above, but in the actual printing process, the shaking of the printed matter occurs due to factors such as mechanical vibration of the printing machine, or the physical properties of the printed object itself. According to this, there is a possibility that pit or swelling may occur during the printing process, so that the determination of the printing failure using the difference of the image is not accurate.

That is, as illustrated in FIGS. 3A to 3C, when an image of a printed matter inputted with respect to an image of a standard model is simply differentially compared, an image of an input printed matter is apparently different from the image of the standard model. Although it did not occur, a large amount of blobs is generated as a large difference occurs in the boundary of the print area as a result of the deformation of the printed matter by external factors. Therefore, a defect determination is made also about the printed matter which has a comparatively favorable printing state.

As a result, in the conventional printing defect inspection apparatus, by comparing only the difference between the image of the standard model and the input printed image, it is difficult to visually distinguish the difference in the printing state between the image of the standard model and the input image. Although it is difficult to determine the degree of difficulty, there is a disadvantage that it is difficult to proceed with error determination of the printed image at high speed.

That is, although the print state of the corresponding input image is good when visually judged, the periphery becomes minutely inconsistent compared to the image of the standardized model due to the slight deformation caused by the minute change around the printed matter. Since it is determined to be, it is disadvantageous that the difficulty of the inspection process is increased due to strict stabilization of illumination around the printed matter to be inspected.

In addition, in the field of manufacturing elastic materials such as clothing and packaging using elastic materials such as cloth, vinyl, and leather, a defect inspection method of printing printed on elastic materials is printed on non-elastic materials. Since it is possible to proceed to the same degree as the inspection method of the printed matter, there is a problem that the frequency of being judged as a printing failure increases, which acts as a factor that inhibits the productivity of the product.

Accordingly, the present invention has been made in view of the above-described conventional circumstances, and an object thereof is to apply a peripheral mask to an image of a printed matter printed on a printed object, and to inspect a defect of a printed material adaptively to various changing printing conditions. To provide a printing defect inspection system using a peripheral mask to be able to perform reliably.

1A and 1B exemplarily illustrate an error state of a printing judgment due to a mismatch of a peripheral portion of a printed matter inspected by a conventional printing failure inspection apparatus;

2A to 2C are views illustrating a detection state of a defective area by a simple difference calculation of a printed image performed by a conventional printing defect inspection apparatus.

3A to 3C are views illustrating an example in which an error is generated by a simple difference calculation of a printed image;

4 is a block diagram showing the configuration of a printing defect inspection system using a peripheral mask according to the present invention;

5 is a diagram illustrating a state of setting an inspection area for a model of an input printed image according to a preferred embodiment of the present invention;

6 is a diagram illustrating a state in which a standard model is generated for comparison with an inspection model of a printed image received from the outside according to an exemplary embodiment of the present invention;

7 is a view showing a state in which similarity calculation is applied to normalized correlations on an image of a printed matter according to a preferred embodiment of the present invention;

8 is a diagram illustrating a state in which a peripheral mask image is obtained by reducing and enlarging an original image of a printed matter according to a preferred embodiment of the present invention;

9 is a view showing a state that the inspection of the printed image according to the preferred embodiment of the present invention by way of example,

10 is a diagram illustrating a state in which a message box indicating an area where an error occurs according to a test result of a printed image is displayed on a management test screen;

11 is a flowchart for explaining a printing defect inspection system using a peripheral mask according to the present invention.

<Description of the symbols for the main parts of the drawings>

10: printer, 12: shooting camera,

14: frame grabber, 16: print inspection device,

18: display monitor, 20: input device,

22: image corrector, 24: image binarizer,

26: model area setter, 28: data storage unit,

30: main processor, 32: production management system.

In order to achieve the above object, according to the present invention, in the inspection system for checking whether or not the printed matter printed on a predetermined printing object by the operation of the printing machine, by taking an image of the printed matter printed by the printing machine Image processing means for image-correcting and binarizing the printed matter image input means received by the printed matter image input means, and a model region which becomes a standard for comparison with the image of the inspection target among the processed image images. Model area setting means for setting an image, generating a peripheral mask image set by varying the size of the standard model image, calculating a difference image between the image of the standard model and the printed image of the inspection target, and calculating the difference image and the surrounding image. By combining mask images, print out the blobs Control means for determining whether the quantity is sufficient, display means for visually displaying an inspection state and a print failure determination state of a printed image to which a peripheral mask is applied, and inspection result data according to a print failure determination result by the control means to the printing machine. It provides a printing defect inspection system using a peripheral mask composed of a data communication means.

According to the present invention configured as described above, it is possible to adaptively inspect the variable printing environment and the material state of the print object by applying a peripheral mask whether or not the print material printed on a predetermined print object through a printing machine, This has the advantage of enabling faster and more reliable inspection of print defects.

Hereinafter, the present invention configured as described above will be described in detail with reference to the accompanying drawings.

That is, Figure 4 is a block diagram showing the configuration of a printing defect inspection system using a peripheral mask according to the present invention.

As shown in FIG. 4, the printing defect inspection system using the peripheral mask according to the present invention includes an interface unit 10a and a printer 10 for printing a printed object on a predetermined printing object, and the printer 10. A photographing camera 12 for capturing an image of a printed matter to be printed through, a frame grabber 14 for acquiring image data of the printed matter photographed from the photographing camera 12 in a standard video data method, and the printer ( 10. A printed matter inspection device 16 having an interface unit 16a for data communication with the printed matter, and inspecting a printing failure state by applying a peripheral mask to an input image of a printed matter obtained from the frame grabber 14, A display monitor 18 for visually displaying the inspection process and the inspection result of the print inspection device 16, and the inspection of the print image image inspected by the print inspection device 16. It consists of an area with an input device 20 for setting by the user input.

In addition, the system of the present invention constitutes a production management system 32 that receives and manages defect detection data according to the print inspection result from the print inspection apparatus 16 through the interface unit 32a.

In the figure, the photographing camera 12 employs an electronic shutter having a speed of, for example, 1/4000 to 1/1000 second (sec) in order to accurately capture the printed matter which is being printed at high speed in the printer 10. The frame grabber 14 is acquired as image data of a standard image type so that the printed image input from the photographing camera 12 has a desired resolution.

In addition, the print inspection apparatus 16 may include an image corrector 22 which performs noise removal and correction processing of a print image input from the frame grabber 14, and an image that optimally binarizes the image correction processed image. Binarizer 24, model region setter 26 for setting the model region of the printed image to be inspected, an input image of the printed matter, an image of the standard model for comparison with the input image, and an image of the standard model A data storage unit 28 storing data of the peripheral mask image obtained by the enlargement and reduction adjustment, and a main processor 30 that determines the occurrence of a defective printout by applying the peripheral mask to the input image of the printout. .

The image corrector 22 performs a smoothing operation to remove noise components generated by an external environment such as lighting, etc. from the printed image obtained from the frame grabber 22, and from the smoothed images, The printing area of the drawing is emphasized using a high frequency filter.

In addition, the image binarizer 24 variably applies Otsu's algorithm (Automatic Selection Of Gray Level For Splitting) according to the state of the printed image through the image corrector 22 to apply only to the text and design region of the corresponding input image. In order to perform a blob analysis on a limited basis, binarization is performed according to appropriate limits.

The model region setter 26 separates only an image of an inspection target from the entire image in order to generate an inspection model for the printed image binarized by the image binarizer 24, as shown in FIG. 5. When the region to be inspected exists in two parts of the first and second inspection regions 40 and 42, the pattern recognition function is applied to the brightness value of the pixel for the left boundary line and the brightness value of the pixel for the right boundary line. By detecting this rapidly varying portion, the positions of the left lines 40a and 42a and the priorities 40b and 42b are obtained, respectively, and the left lines 40a and 42a and the first lines 40b and 42b, respectively. The first and second inspection areas 40 and 42 determined by) are set as the inspection model.

Here, after the left line and the priority are set for the input image, the model region setter 26 designates the entire area of the predetermined printed image through a designation operation by the input apparatus 20 as shown in FIG. 6. Then, the image of the standard model is set, and the characteristic image region designated by the input device 20 is generated as the reference pattern in the image of the standard model.

The reference pattern may be applied to the difference calculation together with the print image of the inspection target and the image of the standard model while the difference processing is performed on the print image of the inspection target and the image of the standard model by the main processor 30. It is supposed to be.

The reference pattern generated by the model area setter 26 is regarded as a normal printing state when visually observed when the entire printed image is compared with the image of the designated standard model and the input image as a whole unit. In the state of difference, as the sum of the minute amounts is generated to a large value and the pattern matching can fail, the partial pattern is compared with the reference pattern designated as the characteristic area to reduce the error in the printing defect inspection.

In addition, the data storage unit 28 stores image data set as an inspection model and peripheral mask data for a standard model region in the printed image, and applies a peripheral mask under a system operating system such as a Windows or Linux program to print poorly. A large storage medium, such as a hard disk drive, is employed to store an application program that is operated to proceed with the inspection.

In the figure, the main processor 30 calculates the difference between the printed image in which the image region is set as the inspection model by the model region setter 28 and the image of the predetermined standard model, and inputs the difference image and input. By combining and processing the image of the peripheral mask set for the image, it is determined whether the printing is defective through the analysis of the blob finally generated. At this time, the main processor 30 applies the reference pattern obtained from the model region setter 26 to the difference calculation when the difference between the printed image of the inspection target and the image of the standard model is obtained.

The main processor 30 determines the inspection object by correlation with respect to the printed images of each inspection model, and obtains a difference between the printed image of the inspection target image and the image of the standard model, and overlaps the plurality of printed images. The normalized correlation is obtained as shown in Equation 2 below.

That is, as shown in Equation 2, as shown in FIG. 7, for the sub-image w (x, y) of the size J × K in the image f (x, y) having an overall size of M × N. Normalized correlation, where S = 0, 1, 2, ..., M-1, t = 0, 1, 2, ..., N-1, and / w is in w (x, y) Is the average value of the pixels, and / f (x, y) corresponds to the average value of f (x, y) in the region coinciding with the current position of w, where the operation of sum is in coordinates common to both f and w Will be done at.

On the other hand, the main processor (30) in order to perform a stable print defect inspection in consideration of the slight changes caused by the expansion or contraction, ambient light caused by the shaking or printing material material characteristics of the printed object, the normal model image Peripheral masks of variable size are applied along the boundaries of the printed area, so that the inspection can be conducted flexibly even if the size of the printed text changes.

That is, as shown in FIG. 8, the main processor 30 executes an application program for generating a peripheral mask and executes a mask extension function and a mask reduction function, thereby normal printing included in the original input image. The expanded printing portion 46 is expanded by n times to generate the expanded printing portion 46, and the normal printing portion 44 is reduced by n times to generate the reduced printing portion 48, while the expanded printing portion 46 is formed. And by combining the reduced printing portion 48, the peripheral mask image 50 is finally generated.

In the same figure, the display monitor 18 displays a printing defect inspection operation to which the periphery mask of the print inspection device 16 is applied so that a worker can visually check, as shown in FIG. By executing a dedicated application program operating under a system operating system such as the above, the blob is extracted by applying a peripheral mask to a plurality of inspection target print images sequentially, and if the blob is larger than a predetermined limit value, the blob is judged as a print defect. do.

Here, the printed matter inspection device 16 is connected to the interface portion 16a of the printer 10 and the interface portion 32a of the production management system 32, the interface portion 16a of which is connected via a serial cable such as RS232C. As a result, the inspection result data according to the result of determining the printing failure of the printed image printed by the printer 10 can be transmitted. As shown in FIG. 10, an error is detected in the inspection region of each printed image. The information on the generated area is shown in detail, for example, "Left 2 error 26.42843".

Next, the operation of the present invention made as described above will be described in detail with reference to the flowchart of FIG.

First, in a state in which the printing machine 10 prints a printed object for a predetermined printing object, the photographing camera 12 photographs the printed matter printed on the printing object passing through the printing machine 10 at high speed, and the frame grabber ( In 14), the printed image captured by the photographing camera 12 is obtained by using a standard image method to have a high resolution (step S10).

In this state, the image corrector 22 of the substrate inspection apparatus 16 performs a smoothing process using a smoothing and a high frequency filter for image correction of the printed image obtained from the frame grabber 14, and performs an image binarizer 24. ) Performs a binarization process to which an appropriate threshold value is applied to the printed image on which the image correction has been performed (step S11).

Next, the model region setter 26 of the printed matter inspection apparatus 16 analyzes pixel values of the lines of the left boundary line and the right boundary line in order to set the model of the inspection target in the printed image, thereby rapidly increasing the pixel values. The variable portion is set to the left line and the priority (step S12). If the left line and the priority are determined to set the inspection model of the printed image, a predetermined print image is selected from among the plurality of printed images through a designated operation by the input device 20. By setting the image as a whole, the image of the standard model is set, and the reference pattern is set by specifying the characteristic region of the image of the standard model by the input device 20 (step S13).

In this state, the main processor 30 calculates the difference between the image of the standard model set by the model region setter 26 and the printed image designated by the model to be inspected by applying an algorithm of image similarity calculation using normalized correlation. (Step S14), the image of the standard model is expanded n times and the image reduced by n times is combined with each other to generate an image of the peripheral mask (step S15).

Next, the main processor 30 performs a calculation process by combining the difference image between the image of the standard model and the printed image of the inspection target calculated in step S14 with the image of the peripheral mask, The remaining blobs are thus extracted (step S16).

On the other hand, the main processor 30 determines whether or not the size of the blob exceeds a preset threshold (step S17), if it is determined that the size of the blob does not exceed the threshold, the inspection It is determined that the printed matter image of the target is normally printed (step S18).

However, if it is determined that the size of the blob exceeds the threshold value according to the determination result of step S17, it is determined that the printed matter to be inspected by the blob is printed as defective and the type and position where the error occurs are calculated. Next, an error occurrence area is displayed on the display monitor 18 (step S19), and the data of the inspection result is produced with the printer 10 through data communication of each interface unit 16a, 10a, 32a. Transfer to management system 32 (step S20).

Therefore, the control means (not shown) for controlling the operation of the printer 10 recognizes the defective state of the printed matter printed on the predetermined print object through the data of the inspection result received from the print inspection device 16 and subsequently In this case, the production management system 32 can utilize the data of the inspection result as data necessary for the production management of the print object.

On the other hand, the present invention having the above-described embodiments can be variously modified and implemented without departing from the spirit and scope of the embodiments.

As described above, according to the present invention, a printing defect is adaptively applied to various variable situations by applying a peripheral mask to a printed image in consideration of a deformation of a print material according to a material of a printed object on which a printed matter is printed or a variable surrounding situation. By inspecting, it is possible to make the inspection process in the printing state quickly and accurately, and not only contribute to the improvement of productivity in the overall manufacturing fixation, but also by applying the peripheral masks to various forms of the printed matter. It has the effect of being able to perform a stable test at all times without receiving.

Claims (4)

In the inspection system for inspecting whether or not the printed matter printed on a predetermined printing object by the operation of the printing machine, Printed image input means for receiving the image of the printed matter printed by the printer and input; Image processing means for performing image correction by binarizing the printed image image inputted by the printed image image input means; Model region setting means for setting an image of a model region which becomes a standard for comparison with an image of an inspection target among the image processed printed images; By generating a peripheral mask image set by varying the size of the standard model image, calculating a difference image between the image of the standard model and the printed image of the inspection target, and combining the difference image and the peripheral mask image, Control means for determining whether or not printing defects by analyzing the occurrence of blobs, Display means for visually displaying an inspection state and a print defect determination state of a printed image to which a peripheral mask is applied; And a data communication means for transmitting inspection result data according to a print defect determination result by the control means to the printer. The method of claim 1, wherein the model region setting means is configured to set a model region by estimating a peripheral boundary of the printed image of the object to be inspected, and generate a standard model image by customizing the image of the set model region as a whole. Print defect inspection system using peripheral mask. The method of claim 2, wherein the model region setting means generates a reference pattern by partially designating a characteristic region in the image of the model region, The control means is a printing defect inspection system using a peripheral mask, characterized in that the calculation is applied to the calculation of the difference when the difference between the print image and the standard model image of the inspection target. The system of claim 1, wherein the control unit combines the image of the standard model image n times and the image of the n times reduced image to calculate a final peripheral mask image. .