JPH0490544A - Method and device for forming of cut out mask - Google Patents

Abstract

PURPOSE:To obtain a correct and desirable contour only by instructing a part of a background section by specifying a part of a region included in the background section of a color picture, dividing the color picture into plural blocks, and obtaining a boundary by analyzing the block at a picture element unit. CONSTITUTION:A part of the region included in the background section in the color picture is specified with an input section 10. On the other hand, the color picture is divided into plural blocks and whether the block belongs to a real object section and a background section or to the region incorporating the boundary between the real object section and the background section is selected. The block of the selected region including the boundary between the the real object section and the background section is analyzed at a picture element unit and the boundary is extracted. For example, the boundary is extracted by conducting a discrimination analysis at every block or by clustering on the whole region. Based on the block selecting it belongs to the background section and the extracted boundary, the cut out mask is formed. Hence only by designating a part of the background section with an operator, the vicinity of a cut out line is selected automatically.

Description

[Detailed description of the invention] [Industrial application field]

The present invention relates to a method and apparatus for creating a cutout mask, and particularly for creating a cutout mask for electronically cutting out and combining color images to obtain a composite image, which is suitable for use in creating a printing plate film. The present invention relates to a method and apparatus for creating a cutout mask suitable for use in.

[Conventional technology]

For example, when printing posters or other commercial products, rather than printing the image of a color transmissive or reflective document as it is, the unnecessary area (background) of the document image is cut out. There are various requests for creating printed matter based on a composite image in which an image of a necessary area (substantial part) is extracted and the actual tree part is combined with a desired background part. A cutout mask is used to extract such real parts and create a composite image. Nowadays, in order to quickly and high-quality cut out a large number of images, a method of creating a cutout mask by electronically processing the original image is being used to create a cutout mask. In addition, various techniques have been proposed for creating cutout masks using electronic image processing (for example, Japanese Patent Application Laid-Open No.
-77065 and Tokuko Sho 63-5745). In the technique disclosed in Japanese Patent Publication No. 64-77065, an input image is displayed on a display screen, an operator indicates the color of the area to be masked while looking at the displayed image, and a cutout mask is automatically created in accordance with the instruction. Further, in the above-mentioned Japanese Patent Publication No. 63-5745, after the operator roughly specifies the area around the cutout line while looking at the image on the display screen, the density of each pixel included in the specified area and the arbitrarily set reference density are calculated. By comparing, the outline of the real # part image is detected and a cutout mask is created.

[Problem to be solved by the invention]

However, in the technique disclosed in Japanese Patent Application Laid-Open No. 64-77065, since the entire image data is binarized based on the specified color, it is often impossible to obtain a correct and desirable outline. Particularly, when the shadow of the real part is in the background part, problems arise in that the shadow part remains or the entire background part remains uncut along the contour line. Furthermore, in the technique disclosed in Japanese Patent Publication No. 63-5745, the operator has to specify the area to be cut out along the contour line, which causes problems such as time consuming and a large amount of work for the operator. The present invention has been made to solve the above-mentioned conventional problems, and an object of the present invention is to provide a method and apparatus for creating a cutout mask that can obtain a correct and desired outline by only specifying a part of the background part. shall be.

[Means to solve the problem]

The present invention provides a method for creating a cropping mask for obtaining a composite image by electronically cropping and compositing color images.
A part of the area included in the background part of the color image is specified, the color image is divided into a plurality of blocks, and based on the specified area and the pixel data of each block, each block is a real part, a background part, or Select which region it belongs to, including the boundary between the real part and the background part, perform a pixel-by-pixel analysis on the block of the selected area including the boundary, extract the boundary, and determine whether it belongs to the background part. The above-mentioned problem is solved by creating a cutout mask based on the selected block and the extracted boundary.6 The present invention also creates a composite image by electronically cutting out and combining color image data. A device for creating a cutout mask for obtaining a cutout mask includes means for specifying a part of an area included in a background part in a color image, and means for dividing the color image into a plurality of blocks and cutting out the specified area and each block. means for selecting, based on pixel data, whether the block belongs to the entity portion 4, the background portion, or a region including a boundary between the entity portion and the background portion; means for analyzing the block pixel by pixel and extracting the boundary; means for creating a cutout mask based on the block selected as belonging to the background portion and the extracted boundary; The above-mentioned problem is also solved by providing a means for storing the created cutout mask.

[Effect]

The present invention provides a method and apparatus for creating a cutout mask in which a part of the area included in the background in a color image is specified. On the other hand, the color image is divided into a plurality of blocks as shown in FIG. Select which of the regions including the boundaries between the background parts belongs. A block of the region including the boundary between the selected entity part and the background part is analyzed pixel by pixel and the boundary is extracted. Extract. A cutout mask is created based on the block selected to belong to the background portion and the extracted boundary. Therefore, the operator can automatically select the area near the cropping line by simply specifying a part of the background, and the boundaries of the selected area can be examined in detail to create a cropping mask. Moreover, a desired contour can be obtained. Therefore, it is possible to obtain a high quality composite image.

[Embodiment 1] Hereinafter, an embodiment of the present invention will be described in detail with reference to the drawings. This embodiment is a cutout mask production apparatus having a configuration as shown in FIG. As shown in Fig. 1, this cutout mask creation device mainly includes an input section 1O for specifying a part of the area included in the background (training area), and a storage area for storing color image data to be cut out. a frame memory 16 as a buffer for displaying the image data in the storage 12 and cropping mask data on the display unit 14; In order to create data, a binarization unit 18 divides the image data into several blocks, and each divided block is divided into a background part, a real part, or a block including a boundary between a real part and a background part. an image dividing unit 20 for selecting whether the boundary belongs to the block, analyzing the block including the boundary pixel by pixel and extracting the boundary; and a display unit 14.2 according to various instructions inputted from the input unit 10.
It includes a control central processing unit 22 for controlling the processing of the digitizing section 18, the image dividing section 20, and the like. Note that, for example, a digitizer or a mouse can be used as the input section 10. Further, a color cathode ray tube (CRT) can be used for the display 114. Next, the operation of the embodiment will be explained. In the cutout mask creation apparatus according to this embodiment, the cutout mask creation process is performed mainly in accordance with the procedure shown in FIG. 2. FIG. 3 shows an example of a state in which an image is processed in each step of this procedure. First, image data is transferred from storage 12 to frame memory 1.
6 (load) (step 1). The image in the frame memory 16 is displayed on the display section 14 as shown in FIG. 3(A), for example. This input image data is taken into the image dividing means 20 and the central processing unit 22 as necessary. Next, a part of the area (training area) included in the background part is specified using the input unit 10 (step 2) 1. For example, the training area is designated as the area indicated by the symbol TA in the image of FIG. 3(A). input. Next, a histogram of the pixel data within the designated training area is taken and analyzed to determine the background color range (step 3). In the training area (
The density y of each color (yellow (Y), magenta (M), cyan (C)) for the pixel in T-A), 1. From the histogram of C, the average density μy, μm, μ for each color
C1 variance σy, σrn, σc, minimum value m1ny,
nir+n+. Find the maximum values 1axy, maxm, and 1axc for 3+nC and #. Next, from these averages .mu.y, .mu.m, .mu.C, etc., the background color range represented by a color solid as shown in FIG. 4(B) is determined on the color space. Here, various methods can be used to obtain the color solid. One of them is the minimum density value of each color, 1iny, lir+Il, 1ine, and the maximum value of 11a.
There is a method in which a fixed area on a color space expanded by α from Xy, 1aXl, and 1aXc is used as a background color area. In this method, for example, the minimum value y-nin and maximum value y-iax of Y in the background color area are expressed by the following equations (1) and (2)%. The background color area can be expanded arbitrarily. Note that the M-axis and C-axis can be determined in the same manner. Another one is the average μy, μm, μC of each color data.
A method can be used in which a region on a color space having a width 2n times the dispersion centering on the background color region is set as the background color region. In this method, for example, the minimum value of Y in the background color area y-1in
, the maximum value y-n+ax can be obtained using the following equations (3) and (4). 'i −111n =μy−n aV −(3)
y −l1ax =μy +n σy−(4) Note that n
is the coefficient. Furthermore, the M-axis and C-axis can also be determined using the same formula. Next, the image data is divided into several blocks, for example as shown in FIG. Label the block "ON", and if the color of the pixels are all outside the background color range, label the block rOFF, and the color of the pixels in the block is the background color. Regardless of the color of the area or other areas, blocks that have a mixture of both will have
Label it "5USPEND" (step 4). An example of labeling blocks in this way is shown in FIG. 3(C). Next, the block labeled rsUsPEND: is analyzed in detail pixel by pixel, and it is determined whether the mask data of the pixel should be set to "ON" or rOFF (step 5). For example, the block is as shown in FIG. 3(D). When making this decision, perform discriminant analysis for each block,
Perform clustering on the entire block,
Alternatively, a method such as setting the range surrounded by an ellipsoid in the color space as the background color region using the maximum likelihood method can be used. Next, the mask data is turned on for the block labeled rONJ in the previous step 4 and the pixels determined to be "ON" in the previous step 5, and a cutout mask image shown by diagonal lines as shown in FIG. 5 is created. is formed and stored in the storage 12 (step 6). Note that, if necessary, the cutout mask data in the storage 12 can be displayed on the screen of the output unit 14, and the cutout mask data can be modified by instructions from the input unit 10. In the embodiments described above, the background color regions for divided blocks and pixels were determined based on the above-mentioned formulas (1) to (4), but pixel discrimination is not limited to this method, and the maximum likelihood method may also be used. The range of the background color area can be set so that the pixels included in the ellipsoid are the background color. [Effects of the Invention] As explained above, according to the present invention, the vicinity of the cutout line can be automatically selected simply by the operator specifying a part of the background, and the selected area can be further detailed and substantiated. By checking the boundaries between the area and the background area, accurately and
The excellent effect of being able to obtain cutout mask data with a desired outline is obtained.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing the device configuration of a cutout mask creation device of the present invention, FIG. 2 is a flowchart showing a mask creation procedure for explaining the operation of the embodiment, and FIG. FIG. 4 is a plan view showing an example of a processed image in the mask creation procedure; FIG. 4 is a diagram for explaining the process of determining a background color area; FIG. FIG. 3 is a plan view showing an example of a cutout mask. 0... Input section, 1 4... Output section, 8... Binarization section, 2... Central processing unit. 20... Image division unit, 2... Storage, 6... Frame memory,

Claims (2)

[Claims] (1) In a method of creating a cropping mask for obtaining a composite image by electronically cropping and combining color images, a part of the area included in the background of the color image is specified, and the color image is divided into a plurality of blocks. selecting whether each block belongs to an entity portion, a background portion, or an area including a boundary between the entity portion and the background portion, based on the designated area and the pixel data of each block; A pixel-by-pixel analysis is performed on blocks in the area including the boundary, the boundary is extracted, and a cutting mask is created based on the block selected to belong to the background area and the extracted boundary. How to create a cutout mask. (2) A device for creating a cropping mask for obtaining a composite image by electronically cropping and compositing color image data, comprising means for specifying a part of the area included in the background of the color image, and the color image data. The image is divided into a plurality of blocks, and based on the specified area and the pixel data of each block, it is determined whether each block belongs to the substance part, the background part, or the area including the boundary between the substance part and the background part. means for selecting; means for analyzing blocks in a region including the selected boundary pixel by pixel and extracting the boundary; a block selected as belonging to the background area; A cutout mask creation device comprising: means for creating a cutout mask based on the created boundaries; and means for storing the created cutout mask.