JPS5821741A - Correcting method for boundary part of image in photoengraving device for television image - Google Patents

Abstract

PURPOSE:To remove the unsightly zigzag parts produced in the boundary parts for images in the case of storing, processing and photoengraving the image of one frame in the stage of making Y, M, BK for printing directly from an NTSC television signal. CONSTITUTION:The timings for writing and reading of a frame memory 24 are such that first the images for the 2 fields of the first half out of repetitions of 4 fields are stored in the memory 24. In succession, the images of the 2 fields of the second half are outputted from an NTSC decoder. The signals thereof are averaged with the feedback signal from a delaying circuit 25 in an averaging circuit 23 and are stored successively in the memory 24. Here, the output of the circuit 25 is the signal delayed by one frame in the circuit 23 and the averaged two signals are exactly the same image. The difference here lies in that the phases of the zigzag components at the boundary in the images are respectively of anti-phases. Therefore only the zigzag components in the boundary parts are eliminated without spoiling the image components at all.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a television image plate making apparatus for producing printing plates for color television images.

It is characterized by creating Y (yellow version, yellow version), 1M (magenta version, sekihan), C (cyan version, gray version), and BK (black version, ink version) for printing directly from the television signal, and is NTSC. This is to improve the deterioration of the 1iIiii image that occurs when one frame of a television signal image is stored, processed, and made into a plate.

Television signals in Japan are composite signals based on the NTSC system developed in the United States. FIG. 1 shows a color bar test signal as an example of a television image. The images include those that have been broadcast, those that have been taken with a camera, etc., and 525 scanning lines (raster) (including blanking) are luminance-modulated to form a surface image. There is no need to go into detail here, but for dual use with conventional black and white TVs, the -3,58MHz phase is treated as π and the phase of -3,58MHz is reversed for each raster. .58MH2 component is taken into consideration so that it does not stand out on the screen.

When watching color television, you don't really notice the defects in the broadcast image, but if you look at the color bar image like the one in Figure 1, you can see the defects. This can be clearly seen in a partially enlarged view of the Sakai area of the signal, as shown in Figure 2, in an image with rich color components and sharp signal rises and falls, such as a color par. . The boundary line is not a single vertical line, but a zigzag pattern as shown in Figure 2.For example, two rasters on the left, two primary rasters on the right, two secondary rasters on the left... It may be arranged in a regular manner like shi, shi,
The zigzag moves slowly from bottom to top as shown by the arrow.

Take a photo of this television image.

At a shutter speed of 1/60 second (equivalent to one field) or 1/30 second (equivalent to one frame), the zigzag shown in Figure 2 can be seen on the photograph. However, when exposed for a long time at a shutter speed of 1/15 seconds (equivalent to 2 frames) or more, the zigzag shown in FIG. 2 does not exist on the photograph. The reason we are not conscious of this zigzag when we watch color television is because it is integrated and canceled out in our heads, just like when taking a long exposure.

FIG. 3 shows a basic block diagram of a television image plate-making device that has already been put into practical use by the applicant.

As a television signal source), VTR (+1 is shown), but as long as it is an NTSC composite signal, the received broadcast image may be output from a television camera or a video disk player. This is a device that directly obtains a printing plate.The original in normal printing is a still image called a color film.Therefore, in the device shown in Figure 3, in order to obtain a still image equivalent to a color film,
Frame memory (3) in the interface (102)
) has been established.

The image stored in the frame memory (3) is VTR
While playing +11, the image specified by the editor can be manually stored by the plate maker, or it can be stored automatically from the frame designation circuit (4).

Images stored as still images in the frame memory (3) can be viewed on a color monitor (5).

The signals used for printing are R (red), G (green), and B (blue).
It is handled as three primary color signals. In order to separate the NTSC composite signal into the three primary color signals, it may be passed through the NTSC decoder or later, but since the readout signal from the frame memory (3) must be slowed down to match the speed of the scanner (103), The explanation will be continued using an example in which the NTSC reader (2) is placed before the frame memory (3).

R, G, H decomposed by NTSC decoder (2)
The three primary color signals are stored in three frame memories 13 respectively.
1 are stored simultaneously. The scanner (103) is equipped with a color computer (6) that converts the three primary color signals of R, G, and H read out from the frame memory (3) into four signals of Y, M, C, and BK for printing, and a color computer (6) that converts the three primary color signals read out from the frame memory (3) into four signals of Y, M, C, and BK for printing. It consists of a part that modulates the brightness and exposes the film, and a rotation control circuit (7).

The film to be exposed (8) is placed in a rotating exposure cylinder (9).
) and rotate the cylinder (9) in the dark. The rotational drive can be performed by the α value of the drive unit with a built-in electric motor, but it also has a rotary encoder.

The pulse signal is supplied as a synchronization signal to the synchronization circuit a3 of the interface (102) through the rotation control circuit (7), and is used as a synchronization signal for reading out the image signal from the frame memory (3). Although not shown, the synchronization circuit includes a platemaking magnification setting circuit.

By controlling the way the signals are read out, the size of the exposed image can be arbitrarily set. For example, if the signal for one scanning line of a television signal is controlled to expose one rotation of the cylinder in a scanner, and each scanning line is sequentially read out and exposed on the film, the scanner (103)
Assuming that the feed rate of the exposure head OD is 500 lines/inch,
The size of the fogged image on the film is approximately 25 x 34.
It is possible to do this.

The clock frequency at the time of reading is set so that distortion or deformation does not occur in the sidelit image.

If one television scanning line is exposed twice in a scanner, the resulting image can be twice as large as the above, 50 x 68 cages. In this way, four printing plates of any size can be obtained.

In this way, the image of one frame (or the image of the l field) stored in the frame and memory IJ-31 is output as four versions of Y, M, C, and BK for direct printing. This image had the above-mentioned drawback.In other words, there was a zigzag at the border of the image with strong changes in nine colors, which made the image difficult to see. .This does not exist on the entire surface, but is noticeable in areas with clear boundaries.1Normally, this is not a problem when making small-sized plates.However, when making large-sized plates, or when the above characteristics are noticeable, The image cannot be used as a printing plate because one frame (or one field) of the image is incompletely decoded as an NTSC signal and stored in memory.

The object of the present invention is to remove the zigzag and unsightly portions that occur at the boundaries of images. FIG. 4 shows a block diagram of a television image boundary correction apparatus according to the present invention. In the present invention, in order to eliminate the above-mentioned drawbacks, two frame memories, a frame memory and a frame memory gradient, are provided in the interface (1G2).

Frame memory Q is a one-frame memory used to store images specified by the editor for platemaking.

The NTSC1 television signal is stored as it is. The frame memory (c) is characterized in that the read frame signal (or field signal) is one still image, and its signal form satisfies the four-field period of the NTSC system. Frame memory 〇! With a car that outputs the output signal of ◇ in this way, after decomposing it into the three primary color signals of R, G, and H using the NTSC decoder (2), it is possible to remove the zigzag at the border in the image, which is a defect, using the frame memory gradient. I can do it. Figure 5 shows the frame memory (
A concrete block diagram of 2) is shown. Since the three signals of frame memory @KE R, G, and H are the same circuit, R
Only the signal circuit is shown and will be explained using this. N
The R signal output from the TSC decoder (2) is sent to the averaging circuit (2).
) and is stored in the frame memory (C), but in order to erase the desired zigzag, it is fed back from the frame memory (C) through the delay circuit (C) to the averaging circuit (C). The principle of the present invention utilizes the four-field repetition of the NTSC system, and the subcarriers in the first two fields and the latter two fields have opposite phases, so that these images are the same still image. It is. FIG. 6 shows a diagram showing the timing of writing to the frame memory (c) and successive output. The first two field images of the four field repetition are stored in the frame memory (c). Subsequently, the second half of the two-frame image is output from the NTSC decoder (2).

As shown in Figure 5, the signal is averaged by the averaging circuit (to) and the feedback signal from the delay circuit (c).
At this time, the output of the delay circuit (A) is stored in the frame memory (C), and the output from the delay circuit (A) is sent to the averaging circuit (C).

The signals are delayed by one frame, and the two signals to be averaged are exactly the same image. What is different at this time is that the phases of the zigzag components at the boundaries in the image are opposite to each other. Namely.

It is possible to remove only the zigzag components at the boundaries without damaging the image components.

The cause of zigzag in the image border is NTS.
Because the YC separation is incomplete in the C decoder (2), that is, the C component (chromaticity component) is mixed into the Y component (luminance component), and the Y component is mixed into the C component.
In the JX circuit, the NTSC composite signal cannot be completely separated into R, G, and H, and this phenomenon appears at the boundary.

If you look at the signal at this boundary with an oscilloscope, you will see that it consists of large ripples with opposite phases for each line. Therefore, in order to remove this ripple component at the border, by averaging the same image signal delayed by one frame as in the present invention, only the ripple component can be removed without any damage to the image signal. .

As mentioned above, the image stored in the frame memory is an image with the zigzag at the border removed, and the liiigI read out from this frame memory (2) in accordance with the rotation of the scanner (103) and made into a plate is a high-quality image. It's a quality image.

Interfuse (102) to obtain a 4-field repetition signal of subcarrier phase in the same image (still image)
However, recent VTRs, especially 1" VTRs, have developed slow and still image playback functions, and the still image playback image, which is a field image, can be of the same high quality as the frame memory output. Now it is possible to transmit this field still playback image through TBC.
If it is formatted into a field repetition signal, the frame memory (c) can be omitted and the still reproduced image signal of VTR (11) can be input directly to the NTSC decoder (2).

The present invention is as described above, and its practical effects are great.

[Brief Description of the Drawings] Figure 1 is an explanatory diagram showing the screen side of a television image, Figure 2 is a partially enlarged view of Figure 1, and Figure 3 is a basic block diagram of a television image plate making device. , FIG. 4 is a block diagram of a television image plate making apparatus used in the method of the present invention, and FIG.
The figure is an enlarged concrete block diagram of the 7-ram memory screen shown in FIG. 4, and FIG. 6 is an explanatory diagram showing the timing of writing and reading. 111-VTR (21-NTSC decoder
(31-・7L/-A memory (6)...Color computer (7)...Rotation control circuit (8)・
... Film to be exposed (9) ... Rotating cylinder for exposure @ ... Synchronous circuit Q]) (a) ... Frame memory patent applicant Toppan Printing Co., Ltd.

Claims (1)

[Claims] Television signals are stored in 7 Rem memory. This is read out, separated into the three primary color signals of R, G, and B by an NTSC decoder, and the three primary color signals are converted into four printing signals of Y, K, C, and BK by a color computer, and these signals are sent from the exposure head to the exposure film. Exposure. In a television image plate making device that creates 4 films for printing (Y, Film, C, and BK), the television signal is stored in the frame memory as an NTSC composite signal, and the still image becomes a 4-field repeating signal of the NTSC system. An NTSC decoder separates the signal into the three primary color signals of RGB, and an averaging circuit averages the first two field signals and the second half two field signals for each signal, thereby eliminating defective signals without damaging image components. 1. A method for correcting image boundaries in a television image plate-making apparatus, the method comprising: removing the image boundaries.