KR101366636B1 - Printed image quality evalution apparatus and printed image quality evalution method - Google Patents

Abstract

The present invention relates to a device and a method for evaluating the quality of a printed image. The device for evaluating the quality of a printed image according to the present invention includes: a binary coding unit converting an evaluation image having a dot area into a binary coded image; a grid extracting unit extracting a grid for separating the dot areas from the binary coded image converted in the binary coding unit; and an image evaluation unit evaluating the binary coded image using the dot area in each grid. Therefore, the concentration of a printed image can be evaluated. [Reference numerals] (110) Evaluation image input unit; (120) Binary coding unit;; (130) Dot area center extracting unit; (140) Grid extracting unit; (150) Valid area selection unit; (160) Image evaluation unit

Description

Print quality evaluation device and print quality evaluation method {PRINTED IMAGE QUALITY EVALUTION APPARATUS AND PRINTED IMAGE QUALITY EVALUTION METHOD}

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a print quality evaluation device and a print quality evaluation method. The present invention relates to a print quality evaluation device and a print quality evaluation method for evaluating the concentration of a print image by a toner or the like.

The brightness of the shadows in the printed result is determined by the number of dots in a certain area, and the image quality is different according to the halftone algorithm that determines the composition of the dots. However, prior to the image quality difference caused by the halftone algorithm, it is basically desired according to the precision of the mechanical, optical and control system such as the optical system of the printer, the shape and size of the toner, the size and amount of ink, the consistency and accuracy of laser or ink ejection, etc. Whether the desired amount of toner or ink is well fixed in the position has a greater influence on the quality of the printed image.

A technique for evaluating the quality of such a printed image is known in Patent Document 1, and the print quality is evaluated using a predetermined pattern in which blacks and white bars are repeated. However, the pattern known in Patent Document 1 has a limit in accurately evaluating the concentration of a printed image.

Patent Document 1: Korean Patent Publication No. 2009-000601

In order to solve the above problems, an object of the present invention is to provide a print quality evaluation device and a print quality evaluation method for evaluating the concentration of a print image by a toner or the like.

In order to achieve the above object, the print quality evaluation apparatus according to the present invention comprises a binarization unit for converting an evaluation image having dot areas into a binary image, and a grid for distinguishing the dot areas of the binary image converted by the binarization unit. A grid extracting section for extracting and an image evaluating section for evaluating a binary image using dot areas in each grid are provided.

It is preferable that each one dot area of the said evaluation image consists of 2 X 2 pattern dots which collected 4 basic dots of a laser beam, or 4 X 4 pattern dots which collected 16 based on DPI (Dots Per Inch).

The binarization unit may calculate a probability distribution from pixel values of light and dark areas of the evaluation image, and perform binarization on light and dark areas based on the calculated probability distribution.

The binarization unit may calculate a reference value for performing binarization on the light and dark areas by using Gaussian distributions of the light and dark areas and respective weighting values based on the histogram of the evaluation image.

The image evaluation unit may change the evaluation frame with a variable parameter in setting the evaluation frame for the dot area in each grid.

The image evaluator may calculate an evaluation value by adding a score when there are black pixels inside the evaluation frame and subtracting the score when there are black pixels in a dot area outside the evaluation frame.

The image evaluator may consider the distance from the evaluation frame in calculating the evaluation value.

In calculating the evaluation value, the image evaluation unit may subtract a score when there are white pixels inside the evaluation frame.

It is preferable that the evaluation image supplied to the said binarization part is an evaluation image photographed enlarged.

The apparatus for evaluating print quality may further include a dot region center extracting unit which extracts a center of each dot region of the binary image converted by the binarization unit.

The dot region center extractor may extract the center of the dot region by a maximum stable extremal region (MSER) detector.

The dot region center extracting unit sets a median value of the center distances between adjacent dots extracted by the MSER detector in the X and Y directions as a representative vector in the direction, thereby determining the area of the dot regions not extracted by the MSER detector. The center can be calculated.

Preferably, the grid extractor extracts a grid by using a straight line connecting the center of gravity to obtain a center of gravity from the centers of four adjacent dot areas.

Preferably, the apparatus for evaluating print quality further includes an effective area selector which selects as a valid area only four grid points of a quadrangle in the grid extracted by the grid extractor.

delete

The image evaluation unit divides the variable parameter into a plurality of values, and selects a parameter value having the highest grade score among the plurality of values as an optimal value.

According to the present invention, there is provided a method for evaluating print quality, comprising: a binarization conversion step of converting an evaluation image having dot areas into a binary image, a grid extraction step of extracting a grid for distinguishing dot areas of the converted binary image, and each grid; By providing an image evaluation step of evaluating a binary image using the dot area within, the above-described object can be achieved.

With the above configurations, the present invention can evaluate the degree of concentration of a print image by toner or the like.

In addition, the present invention can improve the overall process of a printer such as an optical system, a mechanism, a control system by grasping the characteristics of the optical system or the nozzle and the characteristics of the toner or ink from the shape of the dot.

1 is a block diagram of an apparatus for evaluating print quality according to an exemplary embodiment of the present invention.
FIG. 2 is a diagram illustrating an image processing process in the apparatus for evaluating print quality of FIG. 1.
3 is a diagram illustrating a Gaussian distribution of light and dark areas based on a histogram of an evaluation image.
4 is a diagram illustrating a process of obtaining a center of gravity using the centers of four dot areas.
5 is a diagram illustrating a process of setting an evaluation frame for a dot area of a 2 × 2 pattern.
6 is a diagram illustrating a region setting for defining a score function.
7 is a flowchart illustrating a print quality evaluation method according to an embodiment of the present invention.
8 is a diagram illustrating a screen after execution of an image evaluation program implemented according to the present invention.

Hereinafter, a print quality evaluation apparatus and a print quality evaluation method according to an embodiment of the present invention will be described in detail with reference to the accompanying drawings.

1 is a block diagram of an apparatus for evaluating print quality according to an embodiment of the present invention, FIG. 2 is a diagram illustrating an image processing process in the apparatus for evaluating print quality shown in FIG. 1, and FIG. FIG. 4 is a diagram illustrating a Gaussian distribution of light and dark areas based on a histogram of an evaluation image. FIG. 4 is a diagram illustrating a process of obtaining a center of gravity by using centers of four dot areas. FIG. 5 is 2 x 2. FIG. 6 is a diagram illustrating a process of setting an evaluation frame for a dot region of a pattern, and FIG. 6 is a diagram illustrating region setting for defining a score function.

As shown in FIG. 1, the apparatus for evaluating print quality includes an evaluation image input unit 110, a binarization unit 120, a dot area center extraction unit 130, a grid extraction unit 140, an effective area selection unit 150, An image evaluator 160 is included.

The evaluation image input unit 110 acquires an evaluation image in which the printed image printed by the printer is enlarged in order to evaluate the quality of the printed image. The evaluation image acquired by the evaluation image input unit 110 is shown in Fig. 2A. The printed image printed by the printer to evaluate the quality of the printed image has a 2 X 2 pattern or 16 X 4 patterns of one basic dot collected four dots, such as a laser beam, based on dots per inch (DPI). It consists of 4 patterns.

The binarization unit 120 converts the evaluation image having dot areas acquired by the evaluation image input unit 110 into a binary image. Although image quality can be evaluated with a gray image, in the present invention, binarization is performed on the evaluation image to speed up the processing speed and clarify the evaluation.

It may be assumed that the evaluation image acquired by the evaluation image input unit 110 is composed of a bright area and a dark area. The binarization unit 120 calculates a probability distribution of each area from pixel values of the light area and the dark area, and calculates the calculated image. Based on the probability distribution, binarization is performed in light and dark areas. Pixel values in each region are Gaussian distribution,

And assume that the weight of each distribution is P (i). The binarization method used here is to obtain a pixel value when the posterior probabilities of the two regions are equal, and divide them by 0 and 1 based on the values. In other words,

K is obtained, and 0 (black) is given if the pixel value is k or less, and 1 (white) otherwise. For example, when the Gaussian distribution of two regions is predicted based on the histogram of the evaluation image, it is shown in FIG. 3, where the x coordinate of the point where the two distributions meet is a reference value of binarization. When binarization of the evaluation image is performed in this manner, an image as shown in FIG. 2B can be obtained.

The dot region center extractor 130 extracts the center of each dot region of the binary image converted by the binarization unit 120. That is, the dot region center extractor 130 may extract the center of the dot region by a MSS (Maximum Stable Extremal Region) detector. The MSER means a region that is brighter or darker than the surroundings. When the binarization of the evaluation image is performed, the dot region corresponds to the region which is rapidly darker than the surroundings, so that the dot regions can be detected by the MSER detector and the centers thereof can be obtained. In the present invention, the center of the dot region is determined using an MSER detector proposed by J. Matas et al. "Robust wide baseline stereo from maximally stable extremal regions," (British Machine Vision Conference, pp. 384-393, 2002.). Extracted. The image detected by the MSER detector is shown in Fig. 2C.

However, in some dot areas, the center of the dot area may not be detected by the MSER detector. In this case, the dot region center extractor 130 may predict the centers of the remaining dot regions by using the information of the regions detected by the MSER detector. To this end, the dot region center extractor 130 sets the center value of the center distance values with the adjacent dots detected by the MSER detector in the X and Y directions as a representative vector in the direction, and is not extracted by the MSER detector. It is possible to calculate the center of the dot area. That is, the dot region center extractor 130 finds the detected neighboring regions closest to the right and the downward directions with respect to the detected regions, calculates a distance from the neighbors, and among the distances corresponding to the same direction. Take the median and set it as the representative vector in that direction. The dot region center extractor 130 predicts the centers of the clay regions which are not detected by using the representative vector in each direction.

The grid extractor 140 extracts a grid for dividing the dot areas. In order to evaluate the image quality of the dot, the area in which the dot exists must be distinguished. For this purpose, the centers of the dot areas obtained above can be used. As shown in FIG. 4, the grid extractor 140 obtains the center of gravity of the centers of the four dot regions, and obtains a straight line that best connects the centers of gravity to distinguish the regions where the dots exist. An image of the grid extracted by the grid extractor 140 is illustrated in FIG. 2D.

The effective area selector 150 selects only those in which the four grid points of the rectangle exist in the binary image in the grid extracted by the grid extractor 140 as the valid area. A valid area selected by the valid area selecting unit 150 is shown in FIG.

The image evaluator 160 evaluates the binary image only for the dot regions existing in the valid region selected by the effective region selector 150. In the present invention, a case where the diameter of the laser beam is known will be described as an example.

The diameter of the laser beam is 80 mu m, and the shape in which one dot region is formed in a 2 x 2 pattern is shown in FIG. Here, the round circle in one dot area corresponds to the diameter of the laser beam, k shown corresponds to the inverse of the DPI, the inside of the red border means the ideal maximum dot area, and the inside of the black border means the ideal minimum dot area. do. On the other hand, since there may be a difference in the method of expressing the dot area by printer manufacturer or by model, the dot may be fixed only in the area within the black border of FIG. 5, or in some cases, may be fixed on a larger border. Therefore, it is preferable that the image evaluator 160 determine all possible cases while varying d by changing the variable parameter d so that the evaluation frame can be set between the black frame and the red frame. The parameter d can be set from 0 to 18.835 mu m, and the value d when the result is the best can be regarded as the optimum value of the model.

On the other hand, the image evaluator 160 may calculate the evaluation value by adding a score when there are black pixels inside the evaluation frame and subtracting the score when there are black pixels in the dot area outside the evaluation frame. In addition, in calculating the evaluation value, the image evaluator 160 may subtract the score when there is a white pixel inside the evaluation frame, and may also consider the distance from the evaluation frame.

First, the black pixel in the evaluation frame adds a score, and the black pixel in the evaluation frame subtracts the score, but looks at a score function having a score proportional to the linear, square, and square roots as it gets closer to the center. Specifically, when the orange edge of FIG. 6 is a given d value and its inner region is R ^I , the score of the black pixels existing in the dot region may be defined as follows.

Here, di is a distance from the center of a given dot area, and B represents an arbitrary point on the evaluation frame. Therefore, min (B-di) means the minimum value from the black pixel di to the evaluation frame, and the larger value is closer to the center, and smaller value is closer to the evaluation frame. In addition, every score function has one point when there is a black pixel on the evaluation frame.

On the contrary, the black pixels outside the evaluation frame are incorrectly mounted points, and they are given a larger value as they are farther from the evaluation frame and a smaller value as they are closer to the evaluation frame. That is, the following score function is applied to the black pixels in the area R ^o outside the evaluation frame.

Basically, the evaluation function subtracts Equation 11, which is a score given to external points, from Equation 10, which is a score given to points inside the evaluation frame. However, if there are too many white spots inside, it may be desirable to subtract this value in some cases because they act as a negative for the image quality. The combination of score functions considering these is as follows.

Here, xi ^B and xj ^W denote coordinates of the i, j th black and white pixels, respectively, and evaluation mode i is an evaluation function that considers the linear score function of R ^{I and} the linear score function of R ^O as 1: 1. Evaluation mode ii is an evaluation function of a linear score function of R ^{I and} a linear score function of R ^O with a weight of 1: 0.5, and evaluation mode iii is a 1: 1 ratio of the linear score function of R ^I and the square root score function of R ^O. The evaluation mode iv is an evaluation function considering the squared score function of R ^{I and} the square root score function of R ^O as 1: 1, and the evaluation mode v is a score given one point for all pixels in R ^I. It is an evaluation function in which the linear score function of the score and R ^O is different from the weight 1: α.

As a result of comparing the evaluation of each mode according to the above evaluation modes and the subjective evaluation by the evaluator, the similarity between the dual evaluation mode ii and the subjective evaluation by the evaluator was the highest.

The image evaluator 160 divides the variable parameter into a plurality of values, and selects d having a high grade score as an optimal d among the plurality of values.

7 is a flowchart illustrating a print quality evaluation method according to an embodiment of the present invention.

The evaluation image input unit 110 acquires an evaluation image in which the printed image printed by the printer is enlarged in order to evaluate the quality of the printed image (S702).

The binarization unit 120 converts the evaluation image having dot areas acquired by the evaluation image input unit 110 into a binary image (S704). It may be assumed that the evaluation image acquired by the evaluation image input unit 110 is composed of a bright area and a dark area. The binarization unit 120 calculates a probability distribution of each area from pixel values of the light area and the dark area, and calculates the calculated image. Based on the probability distribution, binarization is performed in light and dark areas.

The dot region center extractor 130 extracts the center of each dot region of the binary image converted by the binarization unit 120. For this, an MSER detector may be used (S706).

The grid extractor 140 extracts a grid for dividing the dot areas (S708). In order to evaluate the image quality of the dot, the area in which the dot exists must be distinguished. For this purpose, the centers of the dot areas obtained above can be used.

The effective area selector 150 selects only those in which the four grid points of the rectangle exist in the binary image in the grid extracted by the grid extractor 140 as the valid area (S710).

The image evaluator 160 evaluates the binary image only for the dot regions existing in the valid region selected by the effective region selector 150 (S712). The image evaluator 160 divides the variable parameter into a plurality of values, and selects d having a high grade score as an optimal d among the plurality of values.

8 is a diagram illustrating a screen after execution of an image evaluation program implemented according to the present invention.

The original image is displayed in the left region shown in FIG. 8, and the image processed image is displayed in the middle region. In particular, in the middle region, when the user presses the "ORIGINAL", "BINARY", "Center", "Grid", "Clip", and "Evaluate" buttons shown in the "show" menu in the right region, FIG. The corresponding one of the images of (f) to (f) is displayed.

In addition, it is possible to check whether one dot area is made of a 2 x 2 pattern or a 4 x 4 pattern, and d value for setting an evaluation frame for image evaluation can be manually or automatically selected. If automatically selected, the optimal d value is displayed in the right area after image evaluation. The user can then select an evaluation mode after determining whether to consider the white point in R ^I in the evaluation function.

The grade score by the selected evaluation mode and the automatically calculated optimal d value is displayed at the bottom right.

The foregoing description is merely illustrative of the technical idea of the present invention, and various changes and modifications may be made by those skilled in the art without departing from the essential characteristics of the present invention. Therefore, the embodiments disclosed in the present invention are intended to illustrate rather than limit the scope of the present invention, and the scope of the technical idea of the present invention is not limited by these embodiments. The protection scope of the present invention should be interpreted by the following claims, and all technical ideas within the equivalent scope should be interpreted as being included in the scope of the present invention.

110: evaluation image input unit 110 120: binarization unit 120
130: dot region center extractor 130 140: grid extractor 140
150: effective area selection unit 150 160: image evaluation unit 160

Claims (25)

A binarizer for converting an evaluation image having dot areas into a binary image,
A grid extraction unit for extracting a grid for distinguishing dot areas of the binary image converted by the binarization unit, and
And an image evaluator which evaluates a binary image using dot areas in each grid. The method of claim 1,
Each one dot area of the evaluation image is composed of 2 X 2 pattern dots of 4 basic dots of a laser beam or 4 X 4 pattern dots of 16 based on dots (dots per inch) based on DPI. Device. delete delete The method of claim 1,
And the image evaluating unit may change the evaluation frame with a variable parameter in setting the evaluation frame for the dot area in each grid. 6. The method of claim 5,
And the image evaluator adds a score when there are black pixels inside the evaluation frame, and calculates an evaluation value by subtracting the score when there are black pixels in the dot area outside the evaluation frame. The method according to claim 6,
And the image evaluation unit uses the distance from the evaluation frame in calculating the evaluation value. 8. The method of claim 7,
And the image evaluating unit subtracts the score when calculating the evaluation value when there are white pixels inside the evaluation frame. The method according to any one of claims 1 to 2 and 5 to 8,
And the evaluation image supplied to the binarization unit is an evaluation image which is enlarged and taken. The method according to any one of claims 1 to 2 and 5 to 8,
And a dot region center extracting unit which extracts a center of each dot region of the binary image converted by the binarization unit. 11. The method of claim 10,
And the dot region center extracting unit extracts the center of the dot region by a maximum stable extremal region (MSER) detector. 12. The method of claim 11,
The dot region center extracting unit sets a median value of the center distances between adjacent dots extracted by the MSER detector in the X and Y directions as a representative vector in the direction, thereby determining the area of the dot regions not extracted by the MSER detector. Print quality evaluation device, characterized in that the calculation of the center. The method according to any one of claims 1 to 2 and 5 to 8,
And the grid extractor extracts a grid from a center of four adjacent dot areas and extracts a grid using a straight line connecting the centers of gravity. 14. The method of claim 13,
And an effective area selection unit that selects only those in which the four grid points of the rectangles exist in the binary image as valid areas in the grid extracted by the grid extraction unit. delete delete A binarization conversion step of converting an evaluation image having dot areas into a binary image,
A grid extraction step of extracting a grid for distinguishing dot areas of the converted binary image;
And an image evaluation step of evaluating a binary image using dot areas in each grid. 18. The method of claim 17,
And a dot area center extraction step of extracting a center of each dot area of the binary image converted in the binarization conversion step. 19. The method of claim 18,
In the grid extracting step, a grid is extracted by using a straight line connecting the center of gravity to obtain a center of gravity from the centers of four adjacent dot areas. 20. The method of claim 19,
And an effective area selection step of selecting only those in which the four grid points of the rectangles exist in the binary image as valid areas in the grid extracted in the grid extraction step. 21. The method of claim 20,
The image quality processed by the binarization conversion step, the dot area center extraction step, the grid extraction step, the effective area selection step, and the image evaluation step may be selected or sequentially displayed on a display screen. Assessment Methods. delete delete delete A computer-readable recording medium having recorded thereon a program for executing the print quality evaluation method according to any one of claims 17 to 21.

Images