JPH0246077A - Video processor - Google Patents

Abstract

PURPOSE:To avoid pulling out a picture except objective or missing one part of an objective picture by discriminating a specific code concerning a selective code to select and read video source data written in a memory (color pallet) by picture elements. CONSTITUTION:The output of an encoder 9 is lead out to an output terminal 11 through a change-over switch 10 as a monitor output. A color key signal KY to control the change-over switch 10 is formed by a digital comparator 13 to detect one or plural specific 4-bit codes from the data set selecting signal of a color pallet 8. The data set selecting code is information not related to a level only to select one among 16 pieces. Consequently, a conventional boundary noise generated in the neighborhood of the threshold value of level discrimination is not generated.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention is capable of extracting a specific video source, compositing or switching with other video sources, and writing special effects to a specific video portion into memory [Summary of the Invention] The video source data is specified and read out for each picture element by a selection code, and a specific code in the selection code is discriminated, and the discrimination output is used as a key signal for video processing. Color key operation, which does not require discrimination of color or brightness levels, produces significantly less noise at the boundaries of key areas.

[Conventional technology]

A gate signal called a color key signal or superimpose signal is used to cut out a part of the screen and insert it into another video screen.

Such key signals for screen composition can be simply created based on video synchronization signals, and can be used, for example, to divide the screen into left and right, or to insert images cut into circles or triangles. is formed as a region signal. On the other hand, when cutting out human images, characters, figures, etc., the level (width) of video information (luminance signal, R, G, B component signal, or color difference signal) is compared to distinguish between those above and below a predetermined level. processing to form a key signal.

[Problem to be solved by the invention]

If the luminance signal or R, G, B component signal to be subjected to level discrimination is an analog signal, image deterioration due to the analog comparator is likely to occur along the contour of the cut image. That is, when obtaining the key signal shown in FIG. 8B by discriminating level 5 or higher of the composite video signal as shown in FIG. 8A, good discrimination is possible for parts where the rising and falling edges of the video signal are clear. As shown in the composite screen of FIG. 8C, clean cutting and fitting can be performed based on the key signal.

However, in a portion where the rising and falling slopes of the video signal are gentle, the edges of the key signal fluctuate and fluctuate, resulting in color fringing and color shift at the boundary a of the composite screen. Such image degradation is caused by comparator offset,
It is easily affected by drift, propagation delay, etc., and changes over time.

Even when the granularity signal or R, G, and B signals that are the targets of level discrimination are digitized, the same problems as analog signals occur in principle, especially when it comes to quantization accuracy (ILSB) in the sampling period.
Digital-specific noise corresponding to the noise tends to occur at the border of the embedded screen.

In addition, with the level discrimination method, it is difficult to correctly extract only the intended image, and unintended images may be cut out at the same time, or conversely, the desired image may be cut out with some parts missing. . For example, when performing color key operations on red characters on the screen, if a portion of the red background appears on the screen, that portion is also cut out at the same time, making it impossible to separate it. Further, when extracting a video portion with high brightness, if the brightness level of that portion partially decreases, part of the cut out image will be missing, resulting in an incomplete composite screen.

By making the discrimination level variable, it is difficult to limit the extraction target on the screen or change the extraction area.

In other words, if the discrimination level is lowered in order to correct cropping defects, parts of the screen that are not targeted may be excessively cropped.

In view of the above-mentioned problems, the present invention has extremely little noise generated along the boundaries of the cut and inserted image, can limit the image to be extracted, and accurately cut out only the desired part, and can select any extraction target. It is an object of the present invention to provide a video processing device that provides a degree of freedom in selecting a video processing device.

[Means to solve the problem]

The video processing device of the present invention includes a video source circuit that selects and reads out one of a plurality of types of video source data written in a memory using a selection code input for each pixel to form a video source signal; The present invention is characterized in that it includes a key signal forming circuit that forms a key signal by discriminating a specific code that is being inputted among the selection codes, and uses the key signal to specify an area of the image based on the video source signal.

[Effect]

The selection code is information that specifies the color type for each picture element, and does not have color or brightness level information. These level information occur as the readout output of the memory. Therefore, since no level discrimination is required, there is almost no noise generated along the boundaries of the embedded screen formed based on the key signal, for example.

The target to be extracted by the color key is uniquely determined by the selection code to be discriminated, and there is no possibility that an excess or deficiency of extracted images will occur due to a change in keying conditions.

By changing the code to be discriminated, the sampling target can be changed. By changing the memory contents, the display color of the inset part can be changed, but this has no effect on color key operations.

〔Example〕

Figure 1 shows a CD (compact disc) to which the present invention is applied.
FIG. 2 is a block diagram of a graphics processor. The data format of an optically readable digital audio disc 1 called a compact disc is as shown in Figure 2, where one frame consists of a synchronization signal a and a subcode B.
, PCM audio data c, CIRC error correction code d
It consists of Display data is recorded in the subcode b along with control codes for song intervals and time. Therefore, it is possible to display still images of graphics and characters as well as audio reproduction.

When the disc 1 is played back by the playback device 2, the display data GD in the subcode area is extracted and taken in from the connector 4 to the microprocessor 5 of the graph ink processor 3.The display data is further sent to the graphics controller 6. The pixel data PD is introduced, converted into pixel data, and written into the video memory 7 for one screen.The pixel data PD in the video memory 7 is repeatedly read out at the speed of the video signal and supplied to the color palette 8.

For example, 16 sets of R, G, and B data CL for 16 colors are written in advance in the color palette 8 by the microprocessor 5. The pixel data PD is, for example, a 4-bit data set selection signal. The color palette 8 receives this pixel data as an address, and synchronizes with the dot clock CK (twice the frequency of the color subcarrier fsc) from the graphic controller 6.
RGB component data is derived for each pixel (picture element).

The read RGB component data is supplied to the NTSC or PAL color encoder 9, where the D/A
At the same time, it is converted into a composite color video signal C2 consisting of a luminance signal, a chroma signal and a synchronization signal. The output of the encoder 9 is delivered to an output terminal 11 as a monitor output via a changeover switch 10.

The changeover switch IO is normally fixed to the contact 10a, but when creating a composite screen that superimposes the external video signal EV and CD graphic signal applied from the input terminal 12, the changeover switch IO is fixed to the contact 10a (internal) and 10b ( External) and can be switched within the screen.

The color key signal KY that controls the changeover switch 10 is formed by a digital comparator 13 that detects one or more specific 4-bit codes from the pixel data PD, ie, the data set selection signal of the color palette 8. This comparator 13 is a decoder or a loss detector for discriminating and extracting a specific input code, and can be constructed from a 4-bit input gate/inverter circuit. If necessary, the comparator 13 may be configured with a ROM that receives pixel data PD as an address input, and the code to be detected may be changed, for example, for each screen using a switching signal SEL from the microprocessor 5.

Conventionally, the color key signal KY has been created by distinguishing the levels of the composite color video signal Cv or the component signals R, G, B, but this method has the drawbacks mentioned at the beginning. On the other hand, the comparator 13 in FIG. 1 is not essentially a level discriminator, but
One or more of the codes for selecting one of the GB data sets are discriminated. Although the R, G, and B data sets have color and brightness level information, the data set selection code is simply information unrelated to the level for selecting one out of 16 data sets. Therefore, the conventional boundary noise problem that occurs near the level discrimination threshold essentially does not occur.

FIG. 3 is a functional block diagram illustrating the principle of the color key system of the present invention, and parts corresponding to those in FIG. 1 are given the same reference numerals. The color palette 8 can be considered to be made up of 16 sets of color registers 8-0 to 8-15, each consisting of three registers 15a to 15c for RGB. Sixteen sets of R, G, B color palette data are written into each register 8-0 to 8-15 from the microprocessor 5 through the switch 16 during initial setting.

If each of R and GSB data is 4 bits, it is possible to display 16 colors out of 4096 colors.

The pixel data PD read out from the video memory 7 is
4 to select one of these registers 8-1 to 8-15;
There are 16 types of codes #0 to #15. This register selection code is sent to the register selection signal R3 by the decoder 17.
and is applied to the output side switch 18 of the color palette 8. Therefore, one set of R, G, and B data corresponding to codes #0 to #15 is read out from the switch 18 for each pixel and supplied to the D/A converter of the encoder 9.

On the other hand, the register selection codes #O to #15 are supplied to the digital comparator 13, and a color key signal KY which becomes "1" when the code #0 is input, for example, is formed. When this key signal KY is supplied to the changeover switch 10, only the composite color video signal CV corresponding to the #0 colored pixel is derived from the contact 10a of the switch 10. When video signals of other colors #1 to #15 are generated, the switch 10 switches to the external video signal EV on the side of the contact 10b. Therefore, a composite screen in which a CD graphic display of only color #0 is embedded in the image of the external video signal is displayed on the monitor.

Registers 8-0 to 8-15 of Naori Palette 8 are as follows:
Actually, it is stored in RAM as shown in Fig. 4, and 16
A set of R, G, and B data is written by switching the address, and when selection codes #0 to #15 are designated as an address, one set of them is read out.

FIG. 5 is a block diagram showing an example of a system configuration using the graphic processor 3 of FIG. 1. A video disc player 18 is used as an external video signal source, and a composite screen 20 is displayed on the monitor 21 by combining with display data GD from the disc player 2 (audio disc player). In this case, the image 22 based on the external video signal is a moving image, and the embedded image 23 based on the display data GD is a still image such as lyrics.

FIG. 6 shows the original graphic screen 24 based on data obtained from the subcode area of disk l. This still image screen 24 includes a character 24a and graphics 2.
4b. When creating the display source to be written to disk 1, use selection codes #0 and #1 for the character 24a, and color the graphics 24b using the remaining selection codes #2 to #15. It is decided to do so.

In the graphics processor 8, the character 24
The display data for displaying a is converted into a 4-bit code #0 or #1 in units of pixels forming a character pattern based on font data given to the graphic processor 3 from the host computer 19 in FIG. 8. Display data for displaying the graphics 24b is supplied to the color palette 8 as 4-bit codes #2 to #15.

16 sets of R, G, written in the color palette 8 in advance.
The B color palette data and the above-mentioned font data are written on, for example, a floppy disk FD, and can be loaded through the computer 19 and can be freely rewritten.

The digital comparator 13 discriminates only the #0 or #1 code, regardless of the color or brightness level assigned to the character 24a, and forms a color key signal KY. Therefore, even if the color or brightness of the character 24a and the color or brightness of one of #2 to #15 specified for the graphics 24b are the same or similar, the two will not be confused. Even if the character 24a is a white character and the graphics 24b has a white or very nearly white portion W, a key signal for extracting only the character 24a can be obtained.

Furthermore, even if the coloring form of the screen 24 is changed by rewriting the 16 sets of RGB color palette data in the color palette 8, the color key signal remains unchanged unless the code to be discriminated by the comparator 13 is changed.

Therefore, there is no possibility that a part of the object to be cut out will be missing or unintended cutting will occur due to a change in the coloring conditions. Therefore, it is also possible to alternately change the color of the character 24a, for example, red and white, one second at a time, on the composite screen.

The code to be discriminated by the digital comparator 13 is set to 1 of codes #2 to #15 for the graphics 24b.
It is also possible to set one or more. If the figure of the graphics 24b and its background are further distinguished by code numbers, only that figure can be extracted.

By switching the discrimination code of the digital comparator 13 using the switching signal SEL from the microprocessor 5, the sampling target can be easily changed. Therefore, if the two sets of characters 24a are separately color-coded with two codes #0 and #1, these can be switched and extracted and selectively combined.

FIG. 7 shows an example in which the mixing circuit 25 is used for image synthesis, in which the input graph ink display signal and external video signal are synthesized (superimposed) while changing the mixing ratio with the external video signal. For example, when the key signal is 1'', the level ratio between the graphic and the external video is set to 5:1, and the mixing gain control is performed so that only the external video is derived in other parts.

In the above example, the fitting by switching the switch 10 or the superimposition by the mixing circuit 25 is performed at the analog signal stage, but the external digital component video is It may be combined with the signal. Furthermore, the external video signal may be broadcast video, camera video, or VTR video.
It may also be a digital graphics signal from a computer.

〔Effect of the invention〕

As described above, the present invention provides key signals for video processing by discriminating a specific code for selecting and reading video source data written in a memory (color palette) for each pixel. Therefore, since it does not rely on color or brightness level discrimination as in the past, noise may occur at the edges of the key signal due to the rising and falling slopes of the video information, and the constant of the level discriminator and The keying area does not change due to changes in ambient conditions.In other words, there is no need to extract images other than the target image or lose part of the target image, and the keying target can be uniquely identified by the selection code. Therefore, it is possible to obtain a stable composite image with almost no disturbing noise occurring along the boundaries of the embedded image.

Furthermore, by changing the code to be discriminated, it is easy to change the image to be extracted.

On the other hand, even if the color palette data is rewritten, the keying conditions do not change, so it is possible to freely set the color, brightness, etc. of the extraction target.

[Brief explanation of the drawing]

Fig. 1 is a block diagram of a graphic processor to which the present invention is applied, Fig. 2 is a data format diagram of an optically readable digital audio disc, and Fig. 3 is a functional block diagram of essential parts for explaining the present invention in detail. , Fig. 4 is a block diagram of the color palette memory, Fig. 5 is a system configuration diagram of video synthesis, Fig. 6 is a diagram showing an example of a graphic screen obtained by playing a compact audio disc, and Fig. 7 is a mixing diagram. FIG. 8 is a block diagram showing a synthesis circuit according to the present invention, and FIG. 8 is a diagram showing an example of a time chart and a screen of a conventional level discrimination method. In addition, in the symbols used in the drawings, 1--・-------- Digital audio disc 2-------------------- than more ----・・・
, video memory 8--・---------- color palette 9----------------- color encoder to-------- 1--Switch switch 13--1.--Digital comparator, 20. ...--1--Monitor 22--
····························································−································································································································································································. Agent Masaru Tsuchiya

Claims (1)

[Scope of Claims] A video source circuit that selects and reads out one of a plurality of types of video source data written in a memory using a selection code input for each picture element to form a video source signal; A video processing device comprising: a key signal forming circuit that discriminates a specific code being inputted to form a key signal, and uses the key signal to specify an area of an image based on the video source signal.