Print streak evaluation device and print streak evaluation method

Abstract

To provide a printing streak evaluation apparatus and a printing streak evaluation method which can objectively and accurately evaluate printing-streaks of printing performed on a printing object.SOLUTION: A printing streak evaluation apparatus EA evaluates printing-streaks of printing performed on a printing object, and comprises: image obtaining means 21 that obtains a photographed image in which a print image printed on the printing object is photographed; brightness value obtaining means 22 that obtains brightness values of the photographed image from the photographed image obtained by the image obtaining means 21; roughness parameter calculating means 23 that applies the brightness values obtained by the brightness value obtaining means 22 to a roughness parameter so as to calculate the roughness parameter from the brightness values; skewness calculating means 24 that calculates skewness of the roughness parameter calculated by the roughness parameter calculating means 23; and evaluating means 25 that evaluates the printing-streaks, on the basis of the skewness calculated by the skewness calculating means 24.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a print streak evaluation device and a print streak evaluation method. [Background technology]

[0002] BACKGROUND ART A print streak evaluation method for evaluating print streaks in a print applied to a print object is known (see, for example, Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2020-90612 Summary of the Invention [Problem to be solved by the invention]

[0004] The evaluation method described in Patent Document 1 involves visually evaluating the print streaks on the print applied to the printed object, which has the disadvantage that the evaluation varies depending on the evaluator's subjectivity, making it impossible to objectively and accurately evaluate the print streaks.

[0005] An object of the present invention is to provide a print streak evaluation device and a print streak evaluation method that can objectively and accurately evaluate print streaks in printing applied to a print object. [Means for solving the problem]

[0006] A print streak evaluation device according to one aspect of the present invention is a print streak evaluation device that evaluates print streaks in printing applied to a printing object, and includes: an image acquisition means for acquiring a captured image of a print image printed on the printing object; a luminance value acquisition means for acquiring a luminance value of the captured image from the captured image acquired by the image acquisition means; a roughness parameter calculation means for applying the luminance value acquired by the luminance value acquisition means to a roughness parameter and calculating the roughness parameter from the luminance value; a skewness calculation means for calculating the skewness of the roughness parameter calculated by the roughness parameter calculation means; and an evaluation means for evaluating the print streak based on the skewness calculated by the skewness calculation means.

[0007] In the print streak evaluation device according to one aspect of the present invention, the roughness parameter may be arithmetic mean roughness, root mean square height, maximum valley depth, maximum peak height, maximum height roughness, skewness, or kurtosis.

[0008] In a print streak evaluation device according to one aspect of the present invention, the evaluation means may evaluate the print streaks using a trained model that has been machine-learned using as training data the skewness of a roughness parameter of a teacher image obtained by capturing a print image printed on a learning print object and the evaluation results of the print streaks of the print applied to the learning print object.

[0009] A print streak evaluation method according to one aspect of the present invention is a print streak evaluation method for evaluating print streaks in a print applied to a printing object, and includes the following steps: an image acquisition step for acquiring a captured image of a print image printed on the printing object; a brightness value acquisition step for acquiring a brightness value of the captured image from the captured image acquired in the image acquisition step; a roughness parameter calculation step for applying the brightness value acquired in the brightness value acquisition step to a roughness parameter and calculating the roughness parameter from the brightness value; a skewness calculation step for calculating the skewness of the roughness parameter calculated in the roughness parameter calculation step; and an evaluation step for evaluating the print streaks based on the skewness calculated in the skewness calculation step. [Effects of the Invention]

[0010] According to one aspect of the present invention, roughness parameters are calculated from the brightness values ​​of an image captured of a print image printed on a printing object, and print streaks are evaluated based on the skewness of the roughness parameters. This prevents the evaluation from being affected by the evaluator's subjective judgment, and allows for objective and accurate evaluation of print streaks. [Brief explanation of the drawings]

[0011] [Figure 1] FIG. 1 is an explanatory diagram of a print streak evaluation system according to an embodiment. [Figure 2] FIG. 2 is an explanatory diagram of a print streak evaluation method according to an embodiment. [Figure 3] FIG. 10 is a diagram showing the arithmetic mean roughness and evaluation results. [Figure 4] FIG. 10 is an explanatory diagram of a method for generating a trained model for evaluating print streaks. DETAILED DESCRIPTION OF THE INVENTION

[0012] Hereinafter, an embodiment of the present invention will be described with reference to the drawings. [Device configuration] In Fig. 1, the print streak evaluation device EA is a device that evaluates print streaks on printing performed on a printing medium such as printing paper or printing film, and is configured with a computer such as a personal computer or server. Note that print streaks refer to the occurrence of unintended differences in brightness values ​​in the form of streaks when printing on a printing medium due to differences in ink density or rubbing of the printed surface. The print streak evaluation device EA comprises a storage means 10, a processing means 20, and an operation means 30, and together with a printing means 40, an imaging means 50, and an output means 60 constitute a print streak evaluation system EA1.

[0013] The storage means 10 is configured with a memory, a hard disk, etc., and stores various programs for controlling the print streak evaluation device EA and the print streak evaluation system EA1. In this embodiment, the storage means 10 also stores threshold values ​​of roughness parameters used to evaluate print streaks.

[0014] The processing means 20 is composed of a processor such as a CPU (Central Processing Unit) or a GPU (Graphics Processing Unit), and includes an image acquisition means 21, a brightness value acquisition means 22, a roughness parameter calculation means 23, a skewness calculation means 24, and an evaluation means 25.

[0015] The image acquisition means 21 acquires a captured image of a print image printed on a printing object from the imaging means 50. Examples of the print image include an image, a character, a pattern, an identifier, and a test print pattern.

[0016] The brightness value acquisition means 22 acquires the brightness value of the captured image acquired by the image acquisition means 21 from the captured image.

[0017] The roughness parameter calculation means 23 applies the brightness value acquired by the brightness value acquisition means 22 to the roughness parameter, and calculates the roughness parameter from the brightness value. The roughness parameter is a parameter obtained from the roughness curve that constitutes the profile curve (cross-sectional curve), and is specified in JIS B 0601:2013 and ISO 4287:1997. The roughness parameters include the arithmetic mean roughness Ra, the root-mean-square height Rq, the maximum valley depth Rv, the maximum peak height Rp, the maximum height roughness Rz, the skewness Rsk, and the kurtosis Rku.

[0018] The skewness calculation means 24 calculates the skewness of the roughness parameter calculated by the roughness parameter calculation means 23. The skewness is calculated by multiplying the number of pixels in the captured image by n and the brightness value of the pixel by x i When the average value of the brightness values ​​is x bar and the standard deviation of the brightness values ​​is s, it is calculated using the following formula 1.

[0019]

number

[0020] The evaluation means 25 evaluates the print streaks based on the skewness calculated by the skewness calculation means 24.

[0021] The operation means 30 is composed of a keyboard, an operation panel, a touch panel, a mouse, various switches, a microphone for voice input operation, etc., and is capable of inputting operation signals from various operations into the print streak evaluation device EA and the print streak evaluation system EA1.

[0022] The printing means 40 is composed of an inkjet printer, laser printer, thermal printer, dot printer, offset printing machine, letterpress printing machine, intaglio printing machine, screen printing machine, etc., and is capable of sending and receiving various signals to and from the processing means 20.

[0023] The imaging means 50 is composed of a camera, a camera, a microscope with an imaging function, an imaging sensor, etc., and is capable of transmitting and receiving captured images and various signals to and from the processing means 20.

[0024] The output means 60 is composed of display devices such as displays or panels, and notification devices such as indicator lights or speakers, and is configured to output the evaluation results from the print streak evaluation device EA on a screen or by lighting or sound, etc.

[0025] [Print streak evaluation method] A print streak evaluation method performed in the following procedure shown in FIG. 2 will be described using the print streak evaluation system EA1 equipped with the print streak evaluation device EA described above as an example. First, a user of print streak evaluation system EA1 (hereinafter simply referred to as "user") inputs a signal to start automatic operation to print streak evaluation system EA1 via operation means 30. Next, when the user or a transport means (not shown), such as an articulated robot or belt conveyor, places the printing object in a predetermined position, processing means 20 drives printing means 40 to print on the printing object (step ST11). Thereafter, processing means 20 drives imaging means 50 to capture an image of the print image printed on the printing object (step ST12).

[0026] Next, the image acquisition means 21 acquires a captured image of the print image from the imaging means 50 (step ST13). Then, the brightness value acquisition means 22 acquires the brightness value of each pixel of the captured image acquired in step ST13 (step ST14). In this embodiment, the brightness value acquisition means 22 performs grayscale processing on the captured image and then acquires the brightness value of each pixel.

[0027] Next, the roughness parameter calculation means 23 calculates a roughness parameter from the brightness value acquired in step ST14 (step ST15). In this embodiment, the roughness parameter calculation means 23 calculates the arithmetic mean roughness Ra as the roughness parameter. The arithmetic mean roughness Ra is calculated by the following formula 2, where l is the reference length of the roughness curve and f(x) is the roughness curve.

[0028]

number

[0029] Thereafter, the skewness calculation means 24 calculates the skewness of the roughness parameter calculated in step ST15 (step ST16). Next, the evaluation means 25 evaluates print streaks based on the skewness calculated by the skewness calculation means 24 and outputs the evaluation result to the output means 60 (step ST17). In this embodiment, the evaluation means 25 evaluates print streaks by comparing the calculated skewness with a threshold value stored in the storage means 10. The evaluation means 25 also divides the print streak evaluation into two levels and determines whether print streaks are present (pass) or not (fail). The output means 60 then outputs the print streak evaluation result and notifies the user by displaying it on a display device, turning on an alarm device, or emitting a sound from the alarm device (step ST18).

[0030] If ink is applied uniformly to the printing object so that the area of ​​the ink-free base and the area of ​​the ink-covered portion of the printed image are equal, the variation in the arithmetic mean roughness Ra will be small. In this case, the histogram of the arithmetic mean roughness Ra will approach a normal distribution, as shown in Figure 3, and the skewness value will approach zero. Therefore, if |skewness| is less than or equal to the threshold, it is determined that there are no print streaks.

[0031] On the other hand, even if printing is performed so that the area of ​​the ink-free base is equal to the area of ​​the ink-covered portion, if print streaks occur, the arithmetic mean roughness Ra will vary unevenly. In this case, the histogram of the arithmetic mean roughness Ra will deviate from a normal distribution, as shown in Figure 3, and the skewness value will move away from zero. For this reason, if |skewness| > threshold, it is determined that print streaks are present. Note that the frequency shown on the vertical axis of Figure 3 is the number of pixel rows having that arithmetic mean roughness Ra divided by the number of pixel rows in the entire image.

[0032] In this embodiment, the threshold value for skewness is set to 0.2, and if |skewness|≦0.2, it is determined that there are no print streaks, and if |skewness|>0.2, it is determined that there are print streaks.

[0033] [Evaluation example] A polyvinyl chloride film sample was used as the printing object, and print streaks were evaluated. First, the same color photograph was printed multiple times on the sample using an inkjet printer.

[0034] Next, the quality of the color photograph printed on the sample was visually inspected, and then a portion of the color photograph without streaks and a portion of the color photograph with streaks were imaged at 100x magnification using a digital optical microscope. Each image was then grayscale processed, and the brightness values ​​of each pixel in the image were obtained from the image data as line profiles in the vertical, horizontal, or any direction. The arithmetic mean roughness Ra of the brightness values ​​was then calculated for each line profile, and the skewness of the arithmetic mean roughness Ra was calculated for each image using Equation 1. The calculated skewness and the results of visual inspection of the streaks are shown in Table 1 and Figure 3.

[0035] [Table 1]

[0036] As shown in Table 1 and Figure 3, prints that were evaluated as having no streaks, with |skewness| ≦ 0.2, were confirmed to have no streaks even when visually inspected, while prints that were evaluated as having streaks, with |skewness| > 0.2, were confirmed to have streaks even when visually inspected.

[0037] According to the above-described embodiment, roughness parameters are calculated from the brightness values ​​of an image captured of a print image printed on a printing object, and print streaks are evaluated based on the skewness of the roughness parameters. This means that the evaluation does not change depending on the subjective judgment of the evaluator, and print streaks can be evaluated objectively and accurately.

[0038] As described above, the best configurations, methods, and the like for implementing the present invention have been disclosed in the above description, but the present invention is not limited thereto. That is, although the present invention has been particularly illustrated and described mainly with reference to specific embodiments, those skilled in the art can make various modifications to the above-described embodiments in terms of shape, material, quantity, and other detailed configurations without departing from the scope of the technical idea and purpose of the present invention. Furthermore, the above-disclosed descriptions limiting the shape, material, and the like are provided as examples to facilitate understanding of the present invention and are not intended to limit the present invention. Therefore, descriptions using names of components that are free from some or all of the limitations on shape, material, and the like are included in the present invention.

[0039] For example, the storage means 10 may be built into the print streak evaluating device EA, or may be of a type that is externally attached to the print streak evaluating device EA. When the evaluation means 25 evaluates print streaks using a trained model based on machine learning, the storage means 10 may store the trained model.

[0040] The image acquisition means 21 may acquire the captured image of the printed image printed on the printing object directly from the imaging means 50, or it may not acquire the captured image directly from the imaging means 50, and may acquire the captured image from an image storage means such as a server or database in which the captured images by the imaging means 50 are stored.

[0041] The brightness value acquisition means 22 may acquire brightness values ​​after performing grayscale processing on the captured image, or may acquire brightness values ​​without performing grayscale processing on the captured image.

[0042] The roughness parameter calculation means 23 may calculate a roughness parameter other than the arithmetic mean roughness Ra, and may calculate the arithmetic mean roughness Ra, the root mean square height Rq, the maximum valley depth Rv, the maximum peak height Rp, the maximum height roughness Rz, the skewness Rsk, or the kurtosis Rku.

[0043] The skewness calculation means 24 may calculate the skewness of a roughness parameter other than the arithmetic mean roughness Ra, or may calculate the skewness of the arithmetic mean roughness Ra, the root mean square height Rq, the maximum valley depth Rv, the maximum peak height Rp, the maximum height roughness Rz, the skewness Rsk, or the kurtosis Rku.

[0044] The evaluation means 25 may use a skewness threshold value other than that used in the embodiment for evaluating the presence or absence (pass / fail) of print streaks, or may evaluate print streaks by dividing them into three or more levels depending on the skewness value. For example, if the roughness parameter is the arithmetic mean roughness Ra, then if |skewness|≦0.15, the evaluation may be that there are no print streaks (pass); if 0.15<|skewness|≦0.2, there are almost no print streaks (pass); and if |skewness|>0.2, there are print streaks (fail); or the range of |skewness|=0 to 0.4 may be divided by 40 in increments of 0.1, for example, to evaluate print streaks on a four-level scale, or the skewness value may be output directly as the evaluation result. The evaluation means 25 may change the threshold value of the skewness depending on the type of roughness parameter to be calculated and the ink density of the print applied to the print object.

[0045] The evaluation means 25 may use a trained model that has been machine-learned using skewness as a feature, and evaluate print streaks based on the skewness calculated by the skewness calculation means 24. That is, in the above embodiment, the skewness calculated in step ST16 is compared with a threshold value in step ST17 to evaluate print streaks, whereas the skewness calculated in step ST16 may be input into the trained model in step ST17 to evaluate print streaks.

[0046] The trained model is generated by machine learning using the skewness of the roughness parameters of a training image of a print image printed on a training printing object and the evaluation results of print streaks obtained by visually inspecting the print image printed on the training printing object as training data, and is generated by the following procedure shown in Figure 4.

[0047] First, printing objects of various types and compositions are prepared for learning, and printing is performed on these printing objects for learning (step ST21). Printing may be performed by printing means 40 or by means other than printing means 40, and there are no particular limitations on the shape, size, color, etc. of the print image printed on the printing object.

[0048] Next, the quality of the print image printed on the learning printing object is visually confirmed, and print streaks on the learning printing object are evaluated (step ST22). After that, the print image of the learning printing object is captured and used as a teacher image (step ST23). The image may be captured by the imaging means 50 or by a device other than the imaging means 50. Next, the luminance value of each pixel is acquired from the teacher image captured in step ST23 (step ST24), and a roughness parameter is calculated from these luminance values ​​(step ST25). Then, the skewness of the roughness parameter is calculated as a feature (step ST26). Next, machine learning is performed using the skewness calculated from the teacher image and the evaluation result of the print streaks evaluated in step ST22 as training data, and a trained model is generated in which the skewness of the roughness parameter of the luminance values ​​of the captured image of the print image printed on the printing object is input and the evaluation of the print streaks is output (step ST27). In generating a trained model, the order in which step ST22 and steps ST23 to ST26 are performed is not particularly limited, and steps ST23 to ST26 may be performed after step ST22, or step ST22 may be performed after steps ST23 to ST26. Specific machine learning techniques are not particularly limited and include various known techniques such as neural networks, support vector machines, nearest neighbor methods, random forests, stochastic gradient methods, and kernel approximations.

[0049] The operation means 30 may be configured to be separable from the print streak evaluation device EA, or may be configured to be inseparable. The operation means 30 may or may not be provided in the print streak evaluation device EA or the print streak evaluation system EA1, and if it is not provided, an operation signal from an external input device may be input to the print streak evaluation device EA or the print streak evaluation system EA1.

[0050] The printing means 40 may or may not be provided in the print streak evaluation device EA or the print streak evaluation system EA1, and if it is not provided, printing may be performed on the print object by an external printing device that can communicate with the print streak evaluation device EA or the print streak evaluation system EA1.

[0051] The imaging means 50 may capture the printed image of the printing object in color or in black and white, or may capture a part of the printed image or the entire printed image. The imaging means 50 may or may not be provided in the print streak evaluation device EA or the print streak evaluation system EA1, and if it is not provided, the image acquisition means 21 may acquire an image captured by an external imaging device that can communicate with the print streak evaluation device EA or the print streak evaluation system EA1.

[0052] The output means 60 may or may not be provided in the print streak evaluation device EA or the print streak evaluation system EA1, and if it is not provided, the evaluation results may be output to an external output device that can communicate with the print streak evaluation device EA or the print streak evaluation system EA1.

[0053] The trained model may be one that has been machine-learned using the skewness of the roughness parameter, or may be one that has been machine-learned using the skewness when the arithmetic mean roughness Ra, the root-mean-square height Rq, the maximum valley depth Rv, the maximum peak height Rp, the maximum height roughness Rz, the skewness Rsk, or the kurtosis Rku is used as the roughness parameter.

[0054] There are no particular limitations on the type, material, composition, etc. of the printing object. For example, the printing object may be a substrate surface such as paper, film, adhesive sheet, or pressure-sensitive adhesive sheet, or a print-receiving layer such as a coating layer provided on the substrate surface, and the material may be resin, metal, wood, ceramic, etc.

[0055] The means and steps of the present invention are not limited in any way as long as they can perform the operations, functions, or steps described for those means and steps, and are in no way limited to the components and steps of a single embodiment shown in the above embodiment. For example, the image acquisition means may be any means capable of acquiring a captured image of a print image printed on a printing object, and is not limited in any way as long as it is within the scope of the common general technical knowledge at the time of filing (the same applies to other means and steps). [Explanation of symbols]

[0056] EA: Print streak evaluation device 10...Memory means 20...Processing means 21...Image acquisition means 22...Luminance value acquisition means 23...Roughness parameter calculation means 24...Skewness calculation method 25...Evaluation methods 30...Operation means 40...Printing means 50...imaging means 60...Output means

Claims

1. A print streak evaluation device for evaluating print streaks in a print applied to a print object, comprising: an image acquisition means for acquiring a captured image of a print image printed on the printing object; a brightness value acquiring means for acquiring brightness values ​​of the captured image acquired by the image acquiring means as a plurality of line profiles in any direction; a roughness parameter calculation means for calculating a roughness parameter of the luminance value for each line profile of the luminance value acquired by the luminance value acquisition means; a skewness calculation means for calculating the skewness of the roughness parameters calculated by the roughness parameter calculation means; a print streak evaluation device comprising: evaluation means for evaluating the print streak based on the degree of skewness calculated by the skewness calculation means;

2. 2. The print streak evaluation device according to claim 1, wherein the roughness parameter is an arithmetic mean roughness, a root mean square height, a maximum valley depth, a maximum peak height, a maximum height roughness, skewness, or kurtosis.

3. 3. The print streak evaluation device according to claim 1, wherein the evaluation means evaluates the print streaks using a trained model that has been machine-learned using as training data the skewness of a roughness parameter of a training image obtained by capturing a print image printed on a training printing object and evaluation results of the print streaks of the print applied to the training printing object.

4. A print streak evaluation method for evaluating print streaks in a print applied to a print object, comprising: an image acquisition step of acquiring a captured image of a print image printed on the printing object; a brightness value acquiring step of acquiring brightness values ​​of the captured image acquired in the image acquiring step as a plurality of line profiles in any direction; a roughness parameter calculation step of calculating a roughness parameter of the luminance value for each line profile using the luminance value acquired in the luminance value acquisition step; a skewness calculation step of calculating a skewness of the roughness parameter calculated in the roughness parameter calculation step; and evaluating the print streaks based on the skewness calculated in the skewness calculation step.

Images