JPS58176638A - Preparation of cut mask plate - Google Patents

Abstract

PURPOSE:To extract the outline of an image corresponding to a cut mask, by designating the outlines of an image displayed by color separated image data through a coordinate indicating device. CONSTITUTION:The outline of a displayed image is roughly traced by observing the contracted image of color separated image data of at least one color stored in a memory 1 and a spot image corresponding to a cursor displayed on a monitor video, and moving the cursor of a digitizer 5, and the obtd. approximate data of a desired outline are written in a memory of CPU 3. An outline is extracted from each partial image having an approximate outline nearly in the center, such data are extracted successively one after one, and all the outline data are obtained and written in a recording memory device 6. Then, in a cut mask forming station 4, each image element belonging to the inside of the outline or its outside is converted into a signal value discriminable from other image elements, thus permitting painting out the cut mask.

Description

DETAILED DESCRIPTION OF THE INVENTION A desired contour line or line image of a comic, line drawing, turn, etc. is traced and digital image processing is performed to obtain the contour line or line image. The present invention relates to a method for recording an image having the same shape as the above, and in particular to a method for producing a "cutting mask" in photolithography.

For example, when printing a product catalog, the background of the four-product photo used as the original print image is usually photographed around the necessary double-convex image area, but as a printed product, there is no need to erase this background area. In many cases, only products are displayed.

A cut-out mask is applied for this purpose, by using a film that makes only the necessary image parts transparent and the other parts opaque, and by superimposing this on the original image and duplicating it photographically, A duplicate image with unnecessary background parts removed is obtained.

In addition, when creating a negative image from a positive image, or when using another photographic image as the background area, a cutout mask with the transparent area and light area in the opposite relationship to the above is also used. Ru.

Conventionally, the production of such punched masks has mostly relied on manual work. That is, the outline of one quadrant of the required image is drawn on a transparent sheet using a method similar to tracing work in drafting, and the outside or inside of the outline is filled in with opaque ink. It requires skill and a long time, and is a major obstacle in photoengraving work.

In order to improve this, methods have been used, such as directly filling in opaque ink on photographic film that is a reproduction of the original image, or using so-called beer-off film to save the effort of filling in the film, but this method still remains. ,
@It is not possible to save the trouble of drawing contour lines.

In addition, in addition to setting the background part to a specific color tone and photographing one color during the original image shooting, and duplicating the color image with a color scanner for plate making, it is also possible to extract or erase only the specific color of the background part and create the desired image. There is also a method of obtaining a punched mask, which is disclosed in Japanese Patent Laid-Open No. 17501/1983. However, this method is not generally adopted because it requires assignment at the time of photographing the original image.

In order to overcome the drawbacks and inconveniences of such conventional methods,
Honkuri De Hojin is Tokushuen Sho 53-131195 (Tokukai Shou 53-131195)
No. 5-57846), a method that combines the automatic tracking and recording of image contours by a detection element such as a photoelectric converter and drawing by hand-drawn tracing, and the special Le μSho No. 56-52106, Based on the tracking data obtained in advance by the coordinate indicating device of the digitizer temple and the image signal obtained from the camera device such as an image sensor,
About how to make a mask! vj ed.

The present author has further developed and improved these punching mask manufacturing methods, by using a coordinate indicating device (such as a digitizer tablet) to make a rough outline of the image defect in advance, and then The partial image section of the image,
By sequentially setting based on the contour line and performing digital image processing on each partial image section,
This method is designed to obtain the actual outlines of the two required images, and will be explained below based on Figure Ift11/c.

The blank mask creation station (4) for carrying out the method according to the present invention corresponds to the same or similar number of pixels as the display number (for example, 500 x 500.) of the image display device (8). It is equipped with a memory function that can store partial images, a contour line extraction function, a @contour line connection function, and a function for filling inside or outside the contour area.

The memory capacity of the punching mask creation station (4) is as follows:
Like the casing, it is equipped with a memory block that can capture and store at least two color separation signals, and the data stored here is Evening image display device 18)
Hey, no pressure Will! It can be displayed in color as an image.

Next, the method of producing a punched mask plate by the method of the present invention will be explained.

On the image display device (8), the entire image is compressed by 121 times and displayed from the color-separated image data of at least one color stored in the memory means (1).

This memory means (17) contains, for example, information picked up by a color scanner, color correction, and gradation.

Corrected (or uncorrected) Nb data marked in blue is pre-valued, and its capacitance is determined according to the maximum recording size of one discrimination squadron.

For example, if it is 20 x 30 crn and the length of the recorded version is 200"7;; (500
If it is (ch), then the size of the screen will be 4Kx6 = 24M bytes.

Also, a plurality of memory means (1) are arranged in parallel,
It is also possible to increase the extraction mask creation efficiency by using one more time for creating the intersection b'-vc +i mask.

Pressure 1 [! of this image display device (8)] ! 11 The cursor of the digitizer (5) is moved from the spot image corresponding to the cursor displayed on the monitor screen along with the floor to roughly trace the outline of the image displayed on the image display device (8). . The cursor position at that time is appropriately written into the memory within the CPU (31) as approximate data of the desired contour line.

The approximate data of the contour line is converted by the CPU (a)K so that it corresponds to the address location of the memory.

Next, when constructing the outline shown in center 1, first, a part of the duplicated image is stored in the memory means (1), for example, from the 6th edition memory device in which image signals for the cyan edition are written. That is, the image signal of a partial image corresponding to the same number of pixels as the display pixel number of the image display device (8) is read out and stored in the punching mask creation station (4).

As this partial image, select a part that includes a part of the approximate contour line data approximately in the center, and select each pixel in the part 1...1 image section 111 that is set to include the import and its surroundings. Distinguish whether the tonicity 1 straight is larger or smaller than the average density value of all 1...'' of the partial image area side circles, -h.
Depending on the size relationship, each pixel in the partial image 1, small, 1, and f11 is determined to be rHJ or rLJ.

This partial image section is usually an image display area If, +81
It's a bit small, taking up only one screen, but that doesn't necessarily mean you'll be put off by it.

This kind of determination is based on the 5th year of the Showa era, when the applicant's application is
Similar to that disclosed in No. 6-52106, the CPU
(This is done by the calculation in step 81. As an example is shown in FIG. While duplicating a part of the pixel with
Each time, the above-mentioned classification is performed for each pixel within the horizontal image section.

In this way, approximate data in the partial image t
For all pixels around the IS line, [)iJ,! b Rui is 1
-I-'' is dog-poked, the boundary of its ``H'' and rLJ is traced, and the j za aim corresponding to the current boundary is
This is the contour data of the Il:S fractionation method.

Next, a partial image containing the adjacent portion of the approximate contour data approximately in the center is selected, and contour data for this partial image is obtained in the same manner as described above.

In this way, the approximate contour data is converted into I! By following the table of contents and repeating step 11 above, all the contour line data can be obtained, and these contour edge data are
It is once written into the recording memory device (6).

In this way, when extracting the five-line theta based on the density value of each pixel, the image to be cut out and the weight of the image are close in brightness, so the contour data is automatically extracted. When it is difficult to extract, for example, the cyan version, image signals for other color separation versions such as the magenta version and the yellow version can be read out from the memory means (1) and used. This increases the probability that contour data can be automatically extracted.

Further, when extracting this contour line data, the chroma key method can also be used as described above.

In this case, for each partial image section, the density values of tenocyan, magenta, and yellow for each of all pixels included in the partial image section are read from the corresponding address in the memory device, and each Regarding pixels,
Compare the cyan, magenta, and yellow density values, divide the pixels into pixels corresponding to a pre-specified hue and other pixels, trace the boundaries of these pixels, extract the coordinate values corresponding to the boundaries, and create the contour. Use line data.

In this way, the contour line data extracted for each partial image is divided between adjacent partial images in the partial image section al.
Extraction is performed by overlapping some of the pixels at l.

For this reason, the starting and ending ends of the contour data in each partial image are either overlapped between adjacent partial images, are slightly different from each other, or are completely separated from each other.

Therefore, if the end and start ends of the contour data in adjacent partial images are military envoys, either -
There is no need to perform connection processing just by making the force contour line data the final ring ¥11 line data.

However, as shown in FIG. 3(a), when the end portion and start end portion of each contour line data in adjacent partial images are relatively close to each other, it is necessary to perform a connection process. Therefore, we carefully examine the distances of the coordinate points (marked with circles in Figure 3 (al)) that form each contour line, and select the coordinate points that have the minimum distance (marked with circles in Figure 3 (al)). ) is detected for each contour data and the coordinate points are connected to form the final contour data.

Furthermore, as shown in FIG. 3 (bl), even if the starting and ending portions of the contour line data in adjacent partial images are slightly apart, the joining process is performed twice.

In other words, the coordinate point where the current partial image section and the outline of the approximate data intersect (close enough to Q, the coordinates of the coordinate point group of the contour data extracted within the MiJ partial image section) A coordinate point with a direction that almost matches the direction of the contour line of the approximate data (body and previous partial image section 11)
The coordinate point where the contour line of the approximate data intersects (close to 8, the coordinate point of the coordinate point group of the contour data extracted within this partial image section, and the contour line of the approximate data , and if the distance t between these coordinate points (5) and (S) is small enough, then the coordinate points (5) and (S) are It is also possible to create the final contour data by directly connecting them.

To correct missing or incorrect parts of this contour data,
Each time contour data is extracted for each partial image section, the image display device (8) displays the contour data along with the corresponding partial image.
The partial image section 1 is displayed, visually checked, and if corrections are necessary, corrections are made manually.If there are no corrections required (or if there are no corrections), move on to the next partial image section 1.

If the image you want to cut out has blended into the background of the image and its outline cannot be identified using any of the cyan, magenta, and yellow versions, you can use either the density discrimination method or the chroma key method. However, it is not possible to extract the contour line data, and even after performing the connection process,
It often doesn't connect.

In addition, in the method according to the present invention, in order to eliminate such omission of contour data, the % frequency 56-5210
Correction processing can also be performed in the same manner as disclosed in No. 6.

That is, the entire outline data extracted and written in the recording memory device (6) is stored in the recording memory bag (6).
), by reading it out to the CPU (3) and analyzing it.
The disqualified portion is detected, and then, for example, a cyan image or a color image of the coordinate position corresponding to the missing portion and the outline before and after the missing portion are displayed on the image display device (8). . The input position is indicated by a flicker or the like on the image display device (8)! By tracing the curve corresponding to the disqualified part of Pfr'W with the cursor attached to the displayed digitizer (5), the coordinate data of the disqualified part is added and corrected. This correction process can be performed on all missing parts of the contour data.

Thereafter, the entire image corresponding to the final entire contour line data is displayed on the image display device (8) and visually verified and confirmed.

For example, if it is confirmed that the wrong contour was extracted when extracting contour data, move the cursor attached to the digitizer (51) on the digitizer (51) and (8) When the cursor input position displayed above reaches a predetermined part of the overall image, by inputting the signal from the cursor to the CPU (31),
The coordinate position on the contour data closest to the desired point is detected, and the original image near the coordinates and the circle \11 line [elephant] corresponding to the contour data are displayed on the image display device (8). . Therefore, correction of this portion can be performed in the same manner as the correction process for the missing portion. □ Once the closed outline data of the image corresponding to the desired extraction mask has been extracted, next, at the extraction mask creation station (4), data is drawn inside or outside the outline. The extraction mask is filled in by setting each pixel to which it belongs to a signal value such as a high level signal rHJ or "1" that can be distinguished from other pixels.

An example of such a filling method is shown in FIGS. 4-a to 4-a.

After the closed contour line data is extracted, one pixel clearly outside or inside is converted to L (also f) as shown in Figure 4-a.
iH). Note that it is only necessary to distinguish between inside and outside, and unless the inside is H or open, there is no need to specify it, and you may start from any one point.

From there, the area to the left and right until it straddles the contour line is recognized as the same L (or H) as specified. After straddling the contour line, as shown in Figure 4-b, switch H and L and straddle the next contour line in the same way, but continue until you hit either end of the screen.

In this way, when H and L recognition for one line is completed, each pixel of that line is recognized as Hl or L, which is the same as the starting point, until it straddles the contour line in the vertical direction. However, after crossing the contour line, H and L are replaced and recognized.

Continue this until you hit both the top and bottom edges of the screen.

Thereby, it is possible to recognize whether the screen is H or L from every corner of the screen.

It is also preferable to check the above by monitoring the overall picture.

In this way, the data of the filled-in punched mask image is finally stored in the recording memory device (6).

The above-mentioned cutout mask creation station as a cutout-only station is applied to a layout system that records duplicate images to be compared with a plurality of original images on a single output film, each with a desired duplication magnification and desired arrangement. You can also do that.

Even in this case, for example, two cases are possible.

That is, the force is when applied to a layout system that is an independent mask-dedicated station, and the external force is
This is a case where the layout (including image composition and additional correction) station includes the cutout mask creation station according to the present invention.

FIG. 5 is a block diagram showing an example of the latter layout system to which the blanking mask production station according to the present invention is applied.

In Figure 5, +101 is the same as a normal color scanner.
Digital image signals corresponding to a plurality of original images, which have been subjected to color correction, gradation correction, etc. and converted to a desired reproduction magnification, are inked in each color Y, M, and C, respectively.

A magnetic disk memory device (11) is an interface, (11) is a central processing unit (CPU) such as a microcomputer equipped with a processing program, (13) ) is a magnetic disk memory device into which the final layout-processed image signal is stored, 04) is a digitizer tablet, (to) is an image display device such as a CRT (16+ control circuit, -(17) is a punching mask production station having the same functions as those described above.

In this layout system, first, finish dimensions are specified and the usage area of the magnetic disk memory device Oa+ is set, and then the digitizer tablet lJ4:
For each image data corresponding to each original image,
Provides accurate finished dimensions as specified in the layout and CP
U (121) performs coordinate transformation to a designated position on the finished product, and the image data that has finally been subjected to layout processing is written to the address of the corresponding magnetic disk memory device (18).

At this time, as described above, image data corresponding to the required punching mask may be created each time using the punching mask creation step 17+ and used for layout processing such as trimming. In addition, the extraction mask creating means (1
The image data corresponding to the required punching mask created in step 7) may be read out from a memory device (not shown) and used once.

In this way, all the image data necessary for finishing
In the magnetic disk memory device (when written in the old version, the magnetic disk memory device (13) stores each piece of data at a specified position in an area corresponding to the finished dimensions at the exposure scanning staff density). Dimensions are written in the specified position.

Therefore, when outputting an image onto a film or the like, by outputting the image data written in the magnetic disk memory device (1B) as is, a duplicate image with the desired layout and desired reproduction magnification is recorded.

Furthermore, the method according to the present invention is applicable not only to such a layout system, but also to, for example,
The present invention can also be applied to a simple layout scanner, a so-called convolution scanner, as disclosed in Japanese Patent No. 6-31273.

This compositing scanner is, for example, a Hatram-type scanner, in which the position of the recording head that is sent at a constant speed in the sub-scanning direction is used as a reference, and the recording head and the scanning head for photoelectrically scanning the original image are used to generate duplicate images. By detecting the positional deviations between the recording start point and the scanning start point of the original image, and controlling the feed speed of the scanning head based on these positional deviations and the replication magnification of each replicated image,
One copy image having a desired copy magnification is recorded at each desired position in real time, and the method according to the present invention can be applied to such an apparatus.

That is, by the method according to the present invention, image data corresponding to a punching mask with the missing part filled in is created in advance and written in a memory device, and the image signal read by the scanning head is applied to this punching mask. By controlling based on the corresponding image data, a trimmed duplicate image of a desired shape can be recorded in a desired arrangement at a desired copy magnification.

Furthermore, in the method according to the present invention, image data corresponding to the punching mask is stored in the memory device, and as described above,
It goes without saying that it can be used in another image processing system, and also as a dedicated device for creating a cut-out mask, for recording the exposure of a cut-out mask image on a film.

As described above, according to the present invention, the desired contour #i! inputted by the operator! It is not necessary to trace accurately, just roughly trace, so the cropping efficiency is good.
It is possible to improve the speed of specifying the desired contour line, reduce operator fatigue, and simply enter the necessary data in an interactive format for only the minimum part that requires human judgment. Since the process can be automated, punched mask plates can be created efficiently and in a short time.

In addition, the punched mask plate is made with the same high-precision resolution as that of a color scanner, and contours can be extracted using the density discrimination method or chromakey method, so the desired contour can be formed. Depending on the shape of the image to be used, it is possible to select an efficient discriminative processing power method, and many other effects can be obtained.

[Brief explanation of drawings]

The drawings are for explaining the method of implementing the present invention. Figure 1 is an example of a block diagram for implementing the method according to the present invention, and Figure 2 is an example of a block diagram for implementing the method according to the present invention. FIG. 3 is a diagram showing how to set partial image sections when extracting lines; FIG. Figure 5 is for explaining another embodiment of the method according to the present invention. (1) (to) (19) Memory means (magnetic disk memory device) (2) (11) Interface (3) (Increase central processing unit (CPU) (41 (17)
Without it! Mask creation station (5) (14) Digitizer tablet (6) Recording memory device (7> (15) Control circuit (
81 (161 image display device Fig. 2 Fig. 3 Ca) -1-A

Claims (1)

[Scope of Claims] (1) Fixed data stored in a memory device for writing color-separated image data obtained by scanning an original image with light is displayed on an image display device. Approximately specify the contour line of the image using a coordinate recognition device, and based on the approximately specified contour line, from color-separated image data of at least one color, correspond to a desired punching mask. A method for producing a punched mask plate characterized by extracting a line of Masahiko. (2) Near I↓゛Contour specified like J? A partial image section tooth 1j including fM is sequentially set along the contour line, and each time, the color separation image data value of at least the color of each pixel in the partial image section and the color separation image data value of at least the color of each pixel in the partial image section are set. Claim 1, characterized in that a contour line of an image corresponding to a predetermined punching mask is extracted by comparing the average 11G of the corresponding color separation image data of all pixels in the
1) The method for producing a punching mask plate as described in section 1). (3) Partial image section m1 including the approximately specified contour line
Sequentially, set along the contour #I, each time,
By comparing the three-color separation image data value of each pixel in the partial image section with the three-color separation signal value of a predetermined hue, the outline of the image corresponding to the desired punching mask is extracted. A method for producing a punched mask plate according to claim (1). ! 4+ Sequentially set the partial image section III that includes the approximately specified contour line by cutting the contour line, and each time,
Each pixel in the partial image section has at least one
Compare the color component V image data value of the color with the average value of the corresponding color separation data of all pixels in the relevant partial image area, or compare each pixel in the relevant partial image area. Claim (1) is characterized in that the method selects and executes the comparison between the first three-color separation image data possessed by the computer and the three-color separation signal value possessed by a predetermined hue. How to make a bandit mask. (5) A patent claim characterized in that the extracted contour line is immediately displayed on an image display device together with a color-separated image of at least one color, and the extracted contour line is modified or added as necessary. A method for producing a punching mask according to any one of items (1) to (4).