JPH0722325B2 - Method and device for making cutout mask - Google Patents

Description

Detailed Description of the Invention [Industrial applications]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and an apparatus for producing a cutout mask, and more particularly, to make a cutout mask by displaying color image data on a screen of a display means to create a cutout mask for producing a film for printing plate making. A method and apparatus.

[Prior art]

When creating a film for a printing plate precursor from color image data, a cutout mask may be created for an unnecessary area (for example, a background image area), and a necessary substantial portion may be cut out with the mask. There are various conventional techniques for producing the cutout mask. For example, there is a technique in which input pixel data is displayed on a display screen, and an operator precisely specifies the contour line of the image to be cut out and cuts it out while looking at the image on the screen. There is also a technique in which an operator designates a color of a region (background region) to be masked while looking at an image displayed on a display screen and automatically creates a cutout line from the designation. Also, while looking at the image displayed on the display screen, the operator points out a part of the area to be masked, finds the axis of change in the density of that area, and draws a cutout line according to the density distribution considering the axis of density change. There is a technology for automatic extraction.

[Problems to be Solved by the Invention]

However, in the above-described conventional technique of faithfully cutting out according to the contour line instructed by the operator, it takes time because the operator needs to specify all of the contour lines of the image with a cursor or the like, and the burden on the operator is large. There is a problem. Further, in the technique of automatically creating a cutout line according to a color instruction, when a region of a color close to the background part is inside the substance part to be cut out, there is a possibility that the region may be cut out by mistake. At the same time, there is a problem that even if there is a shadow near the contour, it cannot be cut through well. Further, when an image that cannot be automatically cut out is to be processed fully automatically, there is a problem that it takes a long time for calculation, and the image cannot be cut out correctly, and correction thereof takes time. The present invention has been made to solve the above-mentioned conventional problems, and provides a method and an apparatus for creating a cutout mask, which can create a desired cutout mask with a small number of calculations at high speed and with high accuracy. It is an issue.

[Means for Solving the Problems]

The present invention is a method for displaying color image data on a screen of a display unit to create a cutout mask. In the displayed image data, a range including a contour between a substantial portion and a background portion desired to be cutout is trained. Indicated as an area, for pixels in the training area, a threshold candidate is obtained for each color by an evaluation function calculated based on each color pixel value of the color image data, and the value of the evaluation function of the threshold candidate for each color is set. Based on the selection of the color of the threshold value candidate, the selected threshold value candidate as a threshold value, the pixel data of the color is binarized by the threshold value, based on the background portion of the binarized pixel data, the cutout mask To solve the above problems. Further, according to the present invention, in a cutout mask creating apparatus having a color image data display unit and a calculation unit, the image data displayed on the image display unit is provided with a contour between a substantial portion and a background portion desired to be cut out. Means for indicating a range including the training area, and calculation means for obtaining a threshold candidate for each color by an evaluation function calculated based on each color pixel value of the color image data for pixels in the training area , Based on the value of the evaluation function of the threshold value candidate for each color, while selecting the color of the threshold value candidate, the selected threshold value candidate as a threshold value, the pixel data of the color, the binarization process with the threshold value. , "1",
Regarding the means for converting the binary pixel data of “0” and the binary pixel data in the training area,
By providing means for indicating which of “1” and “0” is the background portion and means for creating a cutout mask based on the binarized pixel data of the background portion Similarly, the above problem is solved.

[Action]

In the present invention, when creating a cutout mask,
For the image data displayed on the screen of the display means, a range including the contour between the substantial part and the background part desired to be cut out is designated as a training area, and for the pixels in the training area, each color pixel of the color image data is specified. A threshold candidate is obtained for each color by an evaluation function calculated based on the value, and the color of the threshold candidate is selected based on the value of the evaluation function of the threshold candidate for each color, and the selected threshold candidate is set as a threshold. And, the pixel data of the color is binarized by the threshold value, based on the background portion of the binarized pixel data,
Create a cutout mask. According to the present invention, since each color pixel data in the training area is targeted, the processing range is smaller than that in the fully automatic clipping. Moreover, not all the pixel data of each color is processed, but the threshold value candidate selected based on the value of the evaluation function of each color pixel data is set as the threshold value, and the pixel data of the color is set to the threshold value of the color. Since the data is digitized, the amount of data to be processed is significantly smaller than that in the conventional case, and thus the calculation load is small. Therefore, the cutout mask can be produced at high speed. Further, by designating a range in which the pixel data of the background part and the pixel data of the substantial part are one color at a time as the training area, the mask can be created more quickly and accurately. Hereinafter, selection of a threshold candidate by an evaluation function calculated based on each pixel value will be described. The evaluation function may be, for example, a document (Journal of the Institute of Electronics and Communications, April 1980 vol, J63-D).
No. 4, No. 4, Nobuyuki Otsu "Automatic threshold selection method based on discrimination and least squares standard"). For example, the density level of monochrome data of a given image is
It is assumed that it has a probability distribution like the histogram of FIG. In this case, the density level k is from 0 to 255, and the probability P (k) of the pixels having the density level with respect to the total number N of pixels is expressed by the following equation (1). P (k) = n (k) / N (1) where n (k) is the number of pixels with a density level of k. The total average level μ _T and the total variance σ _T of pixels are calculated by the following equation (2),
It can be shown by (3). Next, with the level L as a threshold, the cumulative amount ω when classifying into levels 0 to L and L + 1 to 255
(L) and μ (L) are expressed by the following equations (4) and (5). At this time, the occurrence probabilities ω _A and ω _B in the classification from 0 to L and the classification from L + 1 to 255 are expressed by the following equation (6),
It becomes (7). ω _A = ω (L) (6) ω _B = 1-ω (L) (7) Further, the average level is expressed by the following equations (8) and (9). μ _A = μ (L) / ω (L) (8) μ _B = {μ _T −μ (L)} / {1-ω (L)} (9) As a reference indicating the goodness of the threshold value, The ratio η (L) of the inter-cluster variance σ ₀ ² to the total variance σ _T ² is used as in Expression (10). That is, this ratio η (L) is used as the evaluation function of the optimum threshold. η (L) = σ ₀ ² / σ _T ² (10) The variance σ ₀ ² between clusters is expressed by the following equation (11). σ ₀ ² (L) = ω _A (μ _A −μ _T ) ² + ω _B (μ _B −μ _T ) ² = ω _A · ω _B · (μ _A −μ _B ) ² (11) Evaluation function η ( The level L that maximizes L) is set as the optimum threshold value (that is, a threshold value candidate) for the pixel data of a single color.
In this case, the evaluation function η in (L), since the total variance sigma _T ² is constant regardless of the level L, the dispersion between clusters sigma ₀ ²
When (L) becomes maximum, the evaluation function η (L) also becomes maximum. Therefore, the threshold candidate can be determined by calculating the variance σ ₀ ² (L) between the clusters. Here, the variance σ ₀ ² (L) is set to each cumulative value ω (L), μ
When combined with (L), the following equation (12) is obtained. σ ₀ ² (L) = ω (L) {1-ω (L)} × μ (L) / ω (L)-{μ _T −μ (L)} ² / {1-ω (L)} = {[Mu] _{T [} omega] (L)-[mu] (L)} ² / [omega] (L) {1- [omega] (L)} (12) Level L is taken from 0 to 255 and 0 <[omega] (L) <1. The level L having the maximum variance σ ₀ ² (L) in the range is set as the threshold candidate. For the binarized pixel data, if one of “1” and “0” is the background portion,
Since there is no error in making the mask, the speed and accuracy are further improved. Further, the calculation load is reduced as compared with the case of automatically determining which side, and the cutout mask can be produced at a higher speed.

【Example】

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. In this embodiment, color image data for printing plate making of yellow (Y), magenta (M) and cyan (C) having the structure shown in FIG. 2 is displayed on the screen of the display means 14 and the pixel data is displayed. This is a device for making a cutout mask. As shown in FIG. 2, the creating apparatus includes a storage 10 for storing color image data and a storage 10 for storing the color image data.
A frame memory 12 for storing one screen of pixel data in 10 and a color cathode ray tube (CRT) 14 for displaying the image data in the frame memory 12 on the screen,
Regarding the screen displayed by the color CRT 14, a digitizer (or mouse) 16 for instructing a range including the contour between the actual part and the background part desired to be cut out as a training area, and a training area in accordance with the instruction of the digitizer 16 The threshold value candidates of the pixel data of each color plate of C, M, and Y are obtained for the pixels of (1) to (12), and the one having the maximum evaluation function is set as the threshold value among the threshold value candidates of each color. Discrimination analysis means 18 for determining, a binarization means 20 for binarizing the pixel data of the color (the color having the maximum evaluation function) in the training area with the determined threshold value, and binary Digitizer 16 out of the converted image data
And a control CPU 22 for creating master data from the background portion pixel data instructed in. This embodiment has the configuration as described above and creates master data by the procedure shown in FIG. That is, in a state where the color image to be cut out is displayed on the color CRT 14, first, the operator uses the digitizer 16 to display the CRT 14 as shown in FIG.
A range including the contour of the color image on the screen, which straddles the real part to be cut out and the background part, is designated as a training area (step 101). The shape of this training area is arbitrary, and the body portion may be made to go around once as long as the change in color is small. Next, the discriminant analysis unit 18 extracts the pixel data of each color of C, M and Y for the pixels in the training area from the histogram as shown in FIG. 5, for example (1).
~ The threshold candidates ly, lm, and lc are determined using the equation (12) (step 102). Next, with respect to the threshold candidates ly, lm, and lc of each color, the threshold candidate having the largest evaluation function ηymax, ηmmax, ηcmax is determined as the threshold (step 103). In Fig. 5, ηy
max>ηmmax> ηcmax, and the threshold value is that of the Y version. Next, the binarization means 20 determines the pixel data of the color plate (Y plate in the case of FIG. 5) of the color having the largest evaluation function among the three color image data of C, M, and Y. With a threshold
Binarization is performed as shown in FIG. 6 (step 104). Then
Of the pixel data binarized by the operator, “1”,
One of "0" is the background portion is selected and instructed (step 105). For example, in FIG. 6, A is designated as the background portion. Next, based on the pixels included in the designated portion, mask data is created, for example, as shown in FIG. 7, and the mask data is displayed and confirmed on the screen (step 10).
6). As a result, a mask is created within the training area (= cutting area). Next, if another training area is designated, the process returns to step 101 to create a mask, and if not designated, the procedure ends. In the above embodiment, the mask was created for the YMC pixel data, but it is also possible to evaluate the Y, M, and C colors after converting the YMC data into, for example, luminance Y, chromaticity I, and Q data. . That is, the distribution in the YMC space is formed, for example, as shown in FIG.
In the case where the inter-cluster dispersion is insufficient and a clear valley (bottom) of the frequency of each color axis does not appear, it becomes difficult to obtain an accurate threshold value. On the other hand, an accurate threshold value can be obtained by adding a certain kind of coordinate transformation so that a clear valley (bottom) appears in the distribution frequency as shown in FIG. 8 (B). You can The coordinate conversion method is not limited to the above-described luminance Y, chromaticity I, and Q conversion, and any coordinate axis conversion may be used as long as a valley (bottom) can be made to appear and an accurate threshold value can be obtained. Good. In the above embodiment, the Y, M, and C pixel data are processed as they are, but the processing method is not limited to this, and RGB and Y
Considered to have a complementary color relationship (substantially equivalent) with MC, and YM
RGB pixel data can be used instead of C pixel data. Further, the color image data is not limited to that for printing.

【The invention's effect】

As described above, according to the present invention, it is possible to obtain an excellent effect that a cutout mask for image data can be created at high speed and with high accuracy.

[Brief description of drawings]

FIG. 1 is a diagram showing a histogram of density levels for explaining the present invention, FIG. 2 is a block diagram showing a configuration of a cutout mask creating apparatus according to an embodiment of the present invention, and FIG. FIG. 4 is a flow chart showing a mask making procedure in the embodiment, FIG. 4 to FIG. 7 are plan views or diagrams showing image data examples for explaining the operation of the embodiment, and FIG. FIG. 6 is a diagram showing an example of coordinate axis conversion of pixel data in order to obtain a threshold value. 10 ... Storage, 12 ... Frame memory, 14 ... Color cathode ray tube (CRT) as display means, 16 ... Digitizer or mouse, 18 ... Discriminant analysis means, 20 ... Binarization means, 22 ... Control CPU.

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁶ Identification code Internal reference number FI technical display location G06T 7/00

Claims (2)

[Claims] 1. A method of displaying color image data on a screen of a display means to create a cutout mask, wherein a range including a contour between a substantial portion and a background portion desired to be cutout is displayed in the displayed image data. Designated as a training area, for pixels in the training area, a threshold candidate is obtained for each color by an evaluation function calculated based on each color pixel value of the color image data, and the value of the threshold candidate evaluation function for each color Based on the above, the color of the threshold value candidate is selected, the selected threshold value candidate is used as a threshold value, the pixel data of the color is binarized by the threshold value, and the clipping is performed based on the background portion of the binarized pixel data. A method for making a cutout mask, which comprises making a mask. 2. A cutout mask creating apparatus having a color image data display means and a calculation means, the image data displayed on the image display means including a contour between a substantial portion and a background portion desired to be cut out. Means for designating a range as a training area, and a computing means for obtaining a threshold value candidate for each color by an evaluation function calculated based on each color pixel value of the color image data for pixels in the training area, Based on the value of the evaluation function of the threshold candidate for each color, while selecting the color of the threshold candidate, the selected threshold candidate as a threshold, the pixel data of the color, the binarization process with the threshold, Means for converting the binary pixel data of "1" and "0", and the binary pixel data in the training area,
A means for instructing which of "1" and "0" is a background portion and a means for creating a cutout mask based on the binarized pixel data of the background portion are provided. Characteristic cutout mask making device.