JPS58209283A - Video signal inserting system - Google Patents

Abstract

PURPOSE:To realize high picture quality, by using a luminance signal in addition to a color difference signal for forming a key signal for insertion, and by performing naturally the signal switching of a background picture image to be inserted to a foreground picture image according to the contents of the foreground picture image. CONSTITUTION:A foreground luminance signal level difference is added to a background luminance signal component YR by an adder 4 to add preliminary a high space frequency component of the foreground picture image, which floats in the background picture image, to the background picture image after insertion. On the other hand, the foreground luminance signal level difference is lead to an absolute value converting circuit 6. Thereafter, its level is matched with a key signal by a waveform shaping circuit 7, and the absolute level difference of the foreground luminance signal is led to an NAM circuit 9 to perform non-adding mixture with a key signal based on a usual color difference signal sent from a key calculating circuit 8. From the circuit 9, a high resolution picture inserting signal, which consists of the key signal of usual color difference added with the foreground luminance signal level difference component having a high space frequency component, is obtained.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to an image signal embedding method, that is, a so-called chroma key, and in particular, it is a system that enables high-resolution image embedding to be performed in detail to realize a high-quality chroma key. .

Conventional image embedding method of this kind, namely chromakey ta,
Like Shufan, the color of the entire area to be inlaid in the foreground is preferably set to a blue-based color, and the color difference obtained from that area is used. For example, when component encoding is applied to an old colorimetric image, the color difference signal R0-Yo of a single point in the inset area that is colored blue in the foreground image is generated. , Bo-Yo, the distance m on the chromaticity diagram of the color difference signals B-Y, RY of each point of the foreground image
Calculate t from the difference in hue and saturation, and then add adjustable DC voltage 1 to the sum of the beaks multiplied by constants a and b, respectively, and then apply it to the over/flow circuit or underflow circuit. A key signal for insertion is formed by waveform shaping, and a background image is inserted into a required area of the foreground image using the key signal formed only from the color difference signal. However, for example, in the so-called 4:2:2 method regarding the sampling frequency ratio of each component in component encoding of a color image signal, the color difference signal component used for key polarization correction is Since the frequency loss of the inset area is halved to 1, the spatial frequency of the key signal for inset is low, so the foreground image that should be left within the frame of the inset area (C) is Regarding the part, there were drawbacks such as the high spatial frequencies of fine images such as edges in that part were not reproduced, and, for example, parts of a person's hair floating in the blue background of the inset area were not reproduced in detail.

It is an object of the present invention to eliminate the above-mentioned drawbacks of the conventional technology, and to use a luminance signal capable of reproducing high spatial frequencies in addition to the conventionally used color difference signal to form a key signal for embedding. , provides an image signal embedding method that enables high-quality chromakey to be realized by fine-grained switching of image signals by switching signals of a background image that is inserted into a foreground image in a natural manner according to the contents of the entire foreground image. It's about doing.

That is, the image signal insertion method of the present invention aims to insert an area corresponding to the predetermined area of the second color image signal into a predetermined area of the first color image signal, and What is the brightness change within the frame of the predetermined area? weighting the brightness component of the second color image signal of the corresponding area, and then switching the first color image signal of the predetermined area to the second color image signal of the corresponding area; It is characterized by the fact that

The present invention will be described in detail below with reference to the drawings.

First, FIG. 1 shows an example of the configuration of circuit II in the case where image signal embedding according to the present invention is performed between component-encoded color image signals. In the illustrated circuit configuration,
YF tri is the luminance signal component of the color image signal forming the foreground when inset is performed, and YFo is the 4V of a specific point in the middle back region of the foreground color image signal.
YR is the signal level, and YR is the luminance signal component of the background color image signal to be fitted into the blue back region of the foreground image. -10,000, (RY).

and CB-Y)F is the color difference value 4+ of the foreground color image signal
(RY)R and CB-Y)R are color difference signal components of the background color 1jIII image signal. When performing image embedding according to the present invention using each of these signal components, first, the foreground luminance signal stream Y is calculated.

and the specific luminance signal level YFo of the blue background instep and the total subtraction aX
The difference between the two is calculated by calculating the total luminance signal level between each point in the foreground and the specific point in the active hack, and in addition 64, the foreground 14 degree signal level difference is added to the background luminance signal signal. Min YR
The high spatial frequency components of the foreground/song that float in the background image are added in advance to the background image after embedding. , at which point in the foreground image part floating in the background image is switched, the high spatial frequency components of the foreground image part are not lost, and the natural switching is performed smoothly over the details. .

On the other hand, the foreground luminance signal level difference from the subtracter l is guided to the absolute value conversion circuit 6 to calculate the absolute value of the luminance signal level difference. Then, in the waveform shaping circuit 7, the level is matched with the key signal based only on the conventional color difference signal from the key calculation circuit 8, which will be described later, and the foreground luminance No. 1 absolute level difference is calculated as NAMu. 9 and is non-additively mixed with a key signal based on the conventional color difference signal from the 1,000-dimensional calculation circuit 8.

The above-mentioned waveform shaping circuit 7 has input/output characteristics as shown in FIG. 2, for example, and can be easily configured using an IJ-only memory that stores data representing such input/output characteristics. It is something. 2. The output time of this waveform shaping circuit 70 can be linear as shown in the example shown, but the content of the foreground image to be inset, especially the image content around the back area. It is desirable to set the input/output characteristics to be the most wasteful. Generally speaking, it is preferable to set the slope α of the directional curve in the range of 0.2 to 0.4. For example, as shown in FIG. 3, the waveform shaping circuit 7 has multipliers MUL 1 and 2 connected before and after it, and the DC voltage value applied to these multipliers is changed by variable resistors VRI and YR2, respectively. let me,
It is also possible to manually adjust the multiplier factor α to obtain any desired slope α.

As described above, the NAM circuit 9 generates a high-resolution, fine-grained filter that adds a foreground luminance signal level difference component having a good spatial wave number component to the conventional key 2 signal based on the chromaticity signal.
A key signal that allows image embedding is obtained. Key signal overflow/underflow circuit 1o for such resolution change
After shaping the signal waveform, the signal is guided to a low-pass F-wave circuit 11 to smooth the signal waveform, so that so-called soft chroma keying can be performed. Next, the key signal is guided to the key signal distribution circuit 12, which distributes it to a luminance signal switching circuit and a color difference signal switching circuit. After inverting the polarity in inverting circuits 13 and 14, the multipliers 5, 17 and 2o
is supplied as an inversion multiplier, and multiplication processing is performed such that the foreground 4-variant signal component and the background luminance signal component, as well as each foreground color difference signal component and each background color difference signal component, are all differentially increased or decreased and switched. The foreground 4 degree signal component and the background luminance signal component which differentially increase and decrease with respect to each other in this way are
At the same time, each foreground color difference signal component and each background color difference signal component are mixed and switched by adders 16 and 19.
By mixing and switching between the two, it is possible to achieve a high-resolution one-image chroma key.

The image signal embedding method of the present invention can be applied not only to the embedding of component-encoded color image signals described above, but also to the embedding of ordinary color image signals of an analog composite. An example of the circuit configuration of Bazō is shown in FIG. In the illustrated circuit configuration, for example, three primary color image signals R and G from a color turpentine camera are used.
. Regarding the signal Y, the foreground luminance signal level difference similar to that in the configuration example shown in FIG. Hereinafter, in exactly the same manner as in the configuration example shown in the first diagram,
The foreground luminance signal level difference described above and the conventional key signal are expressed as N
By mixing non-daily calculations at the A/M curve 28, a high-resolution key base is formed, and the foreground luminance signal level difference is led to the 7M calculator 24 to create a composite color image as a background. After adding the luminance signals in the signal inset area in advance, the high-resolution key signal described above is applied as is to the multiplier 22 and also applied to the multiplier 25 via the inverting circuit 29, and the foreground color image signal and the background color are The image signals are differentially increased/decreased and combined in an adder 23, and high-resolution chroma keying is performed in detail to obtain a composite color image signal of low-intensity synthesis.

Furthermore, when adding the foreground luminance signal level difference to the background image luminance signal component when inserting the component encoded color image signals into each other using the configuration shown in FIG. If the background image luminance signal component is changed in a so-called modulated form and added to the background image luminance signal component, the background can be embedded into the foreground more naturally.

In order to perform image signal insertion in the manner described above, the first
41 shown! The circuit configuration from the image signal input terminal to the multiplier 2.5 which acts as an image changeover switch in the example is changed as shown in FIG. 5 or 6, respectively.

In the configuration example shown in FIG. 5, the foreground luminance signal component YF and the specific luminance signal level YF in the foreground and back.

and the foreground luminance signal level difference obtained by leading them to the subtracter 30 is supplied to the multiplier 31, and the specific luminance signal level Y
FO'' is supplied to the reciprocal unit 34 which generates 1/x of the input signal to form the reciprocal luminance signal level.The signal is led to the multiplier 35 and multiplied by the background luminance signal component YR to obtain the background image luminance signal level. and is supplied as a multiplier to the foreground luminance signal level difference and multiplied by the foreground luminance signal level difference to obtain a foreground luminance signal level difference of a value corresponding to the magnitude of the ratio, which is led to an adder 32 and added to the background luminance signal component YR. The subsequent signal processing is exactly the same as in the configuration example shown in the first figure.

Due to the above signal processing, the foreground luminance signal level difference becomes modulated by the m original signal level of the background image, and the foreground luminance signal level difference increases in areas where the luminance level of the background image is high. At a low f4 degree level of the background image, the adder for the foreground luminance signal level difference decreases.

Next, in order to simplify the signal processing process in the configuration example shown in FIG. 5, the configuration example shown in FIG. When the value is approximately constant in the range from 0 < α to 1, the coefficient values α of multiple stages corresponding to the values of several stages described above are stored in advance in the coefficient units 35-1 to 35-n. , the optimal coefficient value α depending on the foreground and background image content
By extracting the tooth and supplying it to the multiplier 33 via the changeover switch SW, it is possible to relatively easily obtain the same effect as the configuration shown in FIG.

As is clear from the above description, according to the present invention, regarding the so-called chromakey l/) that performs mutual fitting of color image signals, the key signal for fitting is conventionally formed using the mirror Q) of the color difference signal. Therefore, the spatial frequency of the key signal itself decreased, and high spatial frequency @ components of the foreground image around the blue background area or the area that softened were dropped, making it difficult to reproduce them. On the other hand, the level change of the foreground luminance signal is taken into consideration when forming the key signal for embedding, so that detailed embedding can be performed, and the high spatial frequency components of the foreground image are also added to the background image to be embedded. By adding the values in advance, the high spatial frequency components of the foreground image are prevented from being dropped regardless of the switching position by the key signal, so it is possible to smoothly perform natural high-quality chroma keying according to the image content. . In addition, when soft chroma key 2 is performed by applying low-frequency P waves to the key signal that enables fine inset as described above, although the spatial frequency of the key signal itself is slightly lowered, both the foreground and background Since the high spatial frequency components of the edge portions are complemented between the image signals, the advantages of high-resolution chromakey are not lost.

Note that the image signal insertion method of the present invention can be similarly applied to component-encoded color image signals and analog composite color image signals to achieve the same effect. By adding a simple circuit to a conventional chromakey device, it can be modified to perform high-quality chromakey.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing an example of a circuit configuration of an image signal embedding device according to the present invention, FIG. 2 is a characteristic curve diagram showing an example of input/output characteristics of a waveform shaping circuit in the same configuration example, and FIG. The figure is a circuit diagram showing an example of the configuration of the waveform shaping circuit, FIG. 4 is a block diagram showing another example of the circuit configuration of the image signal insertion device, and FIGS. 5 and 6 are the same as in FIG. FIG. 6 is a block diagram showing some other configuration examples obtained by modifying the configuration example of the image signal insertion device of FIG. 1.80...Subtractor 2, 5.15.17.18.20.22.25.81゜88...Multiplier 3, 4.16.19.23.24.32...Addition Vessel 6
... Absolutely 1 direct chivalry circuit 7 ... Waveform shaping circuit 8.
...Key calculation circuit 9, 28...NAM circuit 10
...Overflow, under 70-control circuit 11...
-Low pass P wave circuit 12...Key signal distribution circuit 13, 14.29...Inversion circuit 21...Color encoder 26...Brightness difference detection circuit 27...Key signal generation circuit 84...・Reciprocal unit 35-1 to 35-n...
・Coefficient unit MULL, 2... Multiplier VRI, 2...
Variable resistor SW... Changeover switch Patent applicant Japan Broadcasting Association No. 1; Xi'7v2 Figure i Figure 3 CDC Figure 4r1 Figure 5 Figure 6

Claims (1)

[Claims] L When inserting a region corresponding to the predetermined singing region of the second color image signal into a predetermined region of the first color image signal, a luminance change component within the frame of the predetermined region of the first color image signal is added to the luminance component of the second color image signal of the corresponding area, and then the first color image signal of the predetermined area is added to the luminance component of the second color image signal of the corresponding area.
The image signal insertion method is characterized by switching to a color image signal.