JP2578431B2 - How to make a picture mask - Google Patents

Description

Description: TECHNICAL FIELD The present invention relates to a method for creating a pattern mask, and more particularly to a method for creating a pattern mask using a computer.

[Conventional technology]

In the field of printing, a technique of extracting only a required picture from an image and using the extracted pattern for printing is often used. in this case,
The part other than the necessary pattern in the image
Requires a mask to cover the surface. For example, consider a case where an image as shown in FIG. 2 (a) is given in the form of a film, and only the coffee cup portion in the image is extracted and used for printing. In this case, the portion of the coffee cup becomes a necessary pattern, and the other portions (including the portion of the cake) become the background. Therefore, as shown in FIG. 2 (b), a mask capable of shielding the hatched portion is created, and the mask covers the background portion of the image shown in FIG. 2 (a).
Only the necessary patterns are extracted.

[Problems to be solved by the invention]

Conventionally, such a mask has been created by manually tracing the outline of a picture. However, there is a problem that such a trace-based mask preparation requires a great deal of labor, and it is difficult to form a precise mask. In particular, when a contour of a tree or the like is a complicated pattern, it is very difficult to faithfully trace the image.

SUMMARY OF THE INVENTION It is an object of the present invention to provide a method for producing a mask for a picture, which can easily produce a precise mask.

[Means for solving the problem]

The present invention divides an image having a pattern into a plurality of pixels,
An image input step of inputting the density value of each pixel as data, a step of defining an outer frame line surrounding the picture in the image, and a step of creating a histogram of the density values of the pixels on the outer frame line, For this histogram, two density areas, a first density area whose frequency exceeds a predetermined number and a second density area whose frequency is equal to or less than the predetermined number, are set. Referring to the density value, it is determined which of the two density areas belongs to, and the image belongs to a closed area of the first attribute composed of a set of pixels belonging to the first density area and to the second density area. Expressed by a second attribute closed region consisting of a set of pixels
For each of the first attribute closed area and the second attribute closed area, it is input whether or not the area is a picture constituting area, and a picture constituting area constituting the picture is extracted from the image. A step of superimposing and displaying the image composed area and the image, a step of inputting information on the specified closed area by the operator based on the display, and newly adding the input closed area to the image composed area, A step of creating a mask based on the updated pattern configuration region, thereby creating a mask of the pattern.

(Operation)

An image is taken into a computer as a set of pixels having density values. Since the computer determines whether the captured pixel is a pixel in the picture or a pixel in the background, the picture can be extracted without the need for a manual operation such as a trace. The above determination is made by taking a pixel on the outer frame line surrounding the picture as a reference pixel in the background and referring to the density histogram of this reference pixel. In addition, the extracted picture configuration area is displayed superimposed on the image. The operator can correct the picture configuration area while watching this display, and can extract a precise picture.

〔Example〕

Hereinafter, the present invention will be described in detail with reference to the illustrated embodiments. FIG. 1 is a flowchart showing a procedure of a mask producing method according to the present invention. The procedure up to step S6 in the procedure described here is the same as the one shown in 1986 by the same applicant as the present application.
The procedure is exactly the same as that disclosed in Japanese Patent Application No. 211947.

Now, consider the case where a mask as shown in FIG. 2B is created for an image as shown in FIG. 2A. First, in step S1, an image is input.
That is, the image as shown in FIG. 2A is divided into a plurality of pixels, and the density value of each pixel is input as data. For example, an image may be optically scanned using an input device such as a scanner, and the density value of each pixel may be captured as digital data by a computer. Generally, this density value is represented by digital data having a power-of-two gradation number. For example, each pixel may be represented by a density value of 0 to 255. The input data of each pixel is stored in a memory in the computer.

Next, in step S2, an outer frame line is defined. This outer frame line is a line surrounding the picture in the image. For example, a rectangular frame as shown by a dashed line in FIG. 2 (c) may be defined as the outer frame line F. Such a definition is displayed on the screen display as shown in FIG. 2 (c) based on the data of each pixel stored in the memory, and the operator uses the input means such as a tablet to display the definition on the display screen. Can be defined as The pixels on the outer frame line become the pixels representing the background portion in the subsequent procedure. Therefore, the outer frame line does not necessarily have to be a quadrangle, nor does it need to be a closed line. However, in order to sample as many pixels as possible representing the background portion over as large a region as possible, it is preferable to make the pixels rectangular as shown in FIG. 2 (c).

Subsequently, in step S3, a histogram of density values is created. That is, the frequency for the density values of all the pixels on the outer frame line F is obtained. This histogram is, for example, as shown in FIG. Generally, such a histogram is a graph having several peaks. FIG. 3 (a) shows a typical histogram which is usually obtained, in which a density region having a frequency of 0 and a density region having a frequency of 1 or more are clearly separated.

Next, in step S4, a density area is set.
That is, a predetermined number a is set for the frequency, and the histogram created in step S3 is divided into a first density area where the frequency exceeds the predetermined number a and a second density area where the frequency is equal to or less than the predetermined number a. Set the density area. For example, in the histogram shown in FIG. 3 (a), if the predetermined number a = 0, as shown in FIG. 3 (b), the first density region indicated by the attribute “0” (frequency exceeding 0) Area) and attribute "1"
And a second density area (area with a frequency of 0) indicated by. Further, for example, in the histogram shown in FIG. 3 (c), the predetermined number a may be appropriately determined so that the peak portion becomes the first density area, and the two density areas may be set as shown in the figure.

Next, in step S5, an attribute “0” closed region consisting of a set of pixels having a density value in the region of the attribute “0” and an attribute consisting of a set of pixels having a density value in the region of the attribute “1” A closed area of “1” is formed. As a result, for example, closed areas A to E as shown in FIG. 2D are formed. Here, the closed area A is a closed area of a background part having the attribute “0”, and the closed area B is a closed area of a necessary picture (coffee cup) part having the attribute “1”. Further, the closed region C is a closed region of the background portion having the attribute “0”. As described above, generally, the closed region of the background portion has the attribute “0”, and the closed region of the required picture portion has the attribute “1”.

However, if the density value distribution of the pixel happens to be similar to the density value distribution of the background portion even though it is a closed region in a necessary picture portion like the closed region D, in the step S5, Attribute "0" is assigned. Also,
Conversely, an unnecessary pattern portion, such as the closed region E, is assigned the attribute “1” although it should be treated as if it were originally a background portion. For this reason, if the attribute of the background part is uniformly processed as "0" and the attribute of the required picture part as "1", inconvenience may occur. Therefore, in step S6, a pattern configuration area is extracted. For example, as shown in FIG. 2 (d), closed areas A, C,
The operator may designate any one point P, Q, R in E by input means such as a graphic tablet. Conversely, the closed areas B and D to be the required picture may be indicated. By this operation, all the closed areas are determined to be either the closed areas to be the background or the closed areas to form the required picture.
Therefore, a mask as shown in FIG. 2B can be automatically created by a computer.

In the above-described embodiment, the present invention has been described by taking an example in which an image is monochrome, but the present invention can be similarly applied to a color image. In this case, the color image
Divide into reference colors such as R, G, B (red, green, blue) or Y, M, C (yellow, magenta, cyan), divide into multiple pixels for each reference color, and Just do it. Subsequently, in steps S2 to S4, the procedure proceeds for each reference color. For example, when R, G, B are used as three reference colors,
A histogram as shown in FIG. 3 (b) is obtained for each of R, G, and B, and for each, a first density area indicated by an attribute “0” and a second density area indicated by an attribute “1” Is set. Next, in step S5, a closed area is formed. At this time, a logical sum of the attributes of all the reference colors of a certain pixel is defined as the attribute of the pixel. That is, pixels belonging to the density area indicated by the attribute “0” for all of the reference colors are defined as pixels having the attribute “0”, and a closed area including a set of pixels having the attribute “0” is defined as a closed area having the attribute “0”. Area, while attribute "1" for any of the reference colors
Pixels belonging to the density area indicated by are defined as pixels having the attribute “1”, and a closed area formed of a group of pixels having the attribute “1” is defined as a closed area having the attribute “1”. In other words, the histograms for each of R, G, and B are superimposed, and the pixels belonging to the density region where all three peaks overlap are the pixels with the attribute “0”, and the other pixels are the pixels with the attribute “1”. Subsequently, if the extraction of the picture configuration area in step S6 is performed in the same manner as the monochrome image, a mask can be created for the color image.

As described above, the procedure up to this point is based on Patent Application No. 21 of 1986.
The technique is the same as the technique disclosed in Japanese Patent No. 1947, but is further improved in the present invention so that the mask can be repaired. In the case of an ideal image, a complete mask as shown in FIG. 2B can be created by the procedure up to step S6, but a complete mask may not actually be obtained depending on the image. For example, as shown in FIG. 4 (a), when the density distribution of the closed region G corresponding to the mouth portion of the coffee cup is close to the density distribution of the background region H,
The background area H and the closed area G are treated as the same area,
As shown in FIG. 6B, the mask completed in step S6 is partially missing. With the method according to the invention, such imperfect masks can be corrected as follows.

First, in step S7, the picture configuration area and the image are superimposed on the display device. This superimposed display can be performed by various methods. For example, in this embodiment, as shown in FIG. 4 (c), of the respective parts constituting the image, the part overlapping the picture forming area is shown in red monotone display. That is, the image in FIG.
The color screen represented by the three primary colors B is displayed on the display device, and the superimposed portion with hatching is represented by the gradation using only the R color element.
This is because, in step S6, after extracting the picture forming region, the average value of the RGB density values of the pixels in this region is set to the R density value, and the G and B density values are set to 0. Should be performed. The hatched area in FIG. 4 (c) is in a state where an image is viewed through a red cellophane. Note that the superimposed display is not limited to such a monotone display, and the superimposed portion can be displayed by hatching, high-luminance display, or various other methods.

The operator can recognize where the incomplete portion of the mask is while watching such superimposed display. Therefore, in step S8, information on the incomplete region is input to compensate for the incomplete closed region. For example, in the example of FIG. 4 (c), the operator inputs information on the closed area G. In this embodiment, this input is performed by inputting the vertex coordinates of the polygon corresponding to the closed area. That is, a cursor is displayed on a display device based on coordinates input from a coordinate input device such as a tablet, and coordinate values are registered according to an instruction of an operator. As shown in FIG. 4 (d), the operator operates the tablet while watching the display screen, and registers several points on the contour of the closed area G. In this example, six points P1 to P6 are registered and 6 points are registered.
Although the closed area G is approximated by a polygon, the number of points to be registered is actually increased to perform more accurate polygonal approximation.

In this way, when the vertex coordinates of the polygon are input,
The computer can determine the closed area by the polygon. By completely treating this closed area as a picture forming area, a complete mask as shown in FIG. 2B is obtained (step S9).

FIG. 5 is an explanatory diagram of a procedure for performing correction by another method.
Now, consider a case where the density distribution of the closed region I is close to the density distribution of the background region H as shown in FIG. 5A, and an incomplete mask is formed as shown in FIG. 5B. . In this case, the superimposed display in step S7 is as shown in FIG.
The portion of the closed region I is not displayed in monotone, and the operator recognizes that a correction input should be made for this region, and makes a correction input in step S8. As described above, the coordinates of the vertices of the polygon can be input. However, the correction input for such a small area can be easily performed as follows. That is, a filled frame J having a size including the small area I to be input is displayed, and the filled frame J is moved on the screen based on an input from a coordinate input device such as a tablet. To do so. Then, the closed area included in the filled frame J may be newly treated as a picture configuration area. This processing is performed by the computer determining whether or not the latter is included in the former based on the coordinate values of the frame of the fill J and the coordinate values of the outline of each closed region. When the operator moves the paint frame J to a position as shown in FIG. 5D using the tablet, the computer newly adds the small area I to the picture composition area. Such a correction method is convenient particularly when the closed region to be corrected is a small region. The operator performs the work of painting the insect-eating area with the brush of the painting frame. When the correction work is completed, a mask is created in step S9 based on the final pattern configuration area, and a complete mask as shown in FIG. 2 (b) is created.

As described above, the example in which the mask is created on the basis of the corrected pattern configuration area data has been described. However, the mask created here is not only used as a mask having a physical substance, but also as data on software. It can also be used as a conceptual mask. As an example of the latter use method, only specific pattern data can be extracted from the image data input in step S1, using the created mask data on the software. That is, raster data as a picture of a coffee cup alone can be generated for a coffee cup from raster data of one screen input from a film as shown in FIG. 2 (a). If such raster data of a single pattern is obtained, it is possible to easily perform a splitting operation of the pattern on the display device.

〔The invention's effect〕

As described above, according to the present invention, in a pattern mask creating method, an image having a pattern is taken into a computer as a set of pixels, and pixels on an outer frame line surrounding the pattern are taken out as reference pixels in a background. The pattern is automatically extracted by referring to the density histogram, and then the pattern composition region and the image are superimposed and corrected so that accurate mask creation can be easily performed. Become like

[Brief description of the drawings]

FIG. 1 is a flowchart showing a procedure of a mask making method according to the present invention, FIG. 2 is an explanatory view of a procedure of an embodiment of a mask making method according to the present invention, and FIG. 3 is made at step S3 in FIG. FIG. 4 and FIG.
The figure is an explanatory diagram showing the operations of steps S7 to S9 in FIG. A to I: closed area, F: outer frame line, J: filled frame, P to R: designated point

 ──────────────────────────────────────────────────続 き Continuation of the front page (56) References JP-A-63-66557 (JP, A) JP-A-58-22886 (JP, A) JP-A-58-166348 (JP, A) JP-A-58-228 176638 (JP, A) JP-A-60-117375 (JP, A)

Claims (7)

(57) [Claims] An image input step of dividing an image having a picture into a plurality of pixels and inputting a density value of each pixel as data; a step of defining an outer frame line surrounding the picture in the image; Creating a histogram for the density values of the pixels on the outer frame, and for the histogram,
Setting two density areas, a first density area whose frequency exceeds a predetermined number, and a second density area whose frequency is equal to or less than the predetermined number; It is determined which of the two density areas belongs to the image by referring to the closed area having a first attribute including a set of pixels belonging to the first density area and the second attribute.
And a second attribute closed region consisting of a set of pixels belonging to the density region of the first and second attributes, and for each of the first attribute closed region and / or each of the second attribute closed regions, Entering whether or not it is an area to configure, extracting a pattern configuration area that configures a pattern from the image, and displaying the image and the pattern configuration area in a superimposed manner, based on the display Inputting information about the closed area specified by the operator, and adding the input closed area to a new pattern configuration area, and creating a mask based on the updated pattern configuration area. How to make a mask for a picture. 2. A color image is divided into a plurality of pixels for each predetermined reference color, two density areas are set for each of the reference colors, and all of the reference colors belong to the first density area. A closed region consisting of a set of pixels is defined as a closed region of a first attribute, and a closed region consisting of a set of pixels belonging to a second density region with respect to any of the reference colors is defined as a closed region of a second attribute. 2. The method for producing a picture mask according to claim 1, wherein: 3. The pattern according to claim 1, wherein the density value of each pixel is represented by digital data having a power-of-two gradation number and input. How to make a mask. 4. The method according to claim 1, wherein a predetermined number relating to the frequency of the histogram is set to zero. 5. The picture according to claim 1, wherein the input of the information on the closed area is performed by inputting the vertex coordinates of a polygon corresponding to the closed area. How to make a mask. 6. The information processing apparatus according to claim 1, wherein the input of the information on the closed area is performed by designating another closed area including the area. How to make a picture mask. 7. The pattern according to claim 1, wherein the superimposed display of the image and the pattern constituting area is performed by displaying a superimposed portion of the image in monotone. How to make a mask.