JPH0686307A - Muse decoder with color noise reducer - Google Patents

Abstract

PURPOSE:To perform noise reduction to chrominance signals utilizing exact correlation and further, to effectively perform the noise reduction for the noise converted to a conspicuous lowfrequency area by time expansion. CONSTITUTION:An adaptive type color noise reducer 31 provided in the rear stage of an adaptive mixing processing means 20 performs the noise reduction by utilizing the fact that line sequential color difference signals lead as the chrominance signals are correlated between two frames. At this time, by motion signals indicating the motion of video images from a motion detection processing means 15, a noise reducing amount is made small in an area with the large motion of the video images, the noise reducing amount is made large in the area with the small motion of the video images and the noise reduction of the linear sequential color difference signals is adaptively performed.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a MUSE decoder in a high-definition receiver, and more particularly to a MUSE decoder with a color noise reducer equipped with a color noise reducer that effectively removes noise of color signals.

[0002]

2. Description of the Related Art M is one of the high definition broadcasting systems.
The USE (Multiple Sub-Nyquist Sampling Encoding) method was developed by the Japan Broadcasting Corporation (NHK). About this MUSE method, Ninomiya et al.
Development of system ”NHK Technical Research, Showa 62, Vol. 39, No. 2

This MUSE system is for band-compressing a wide band high-definition signal using sub-sampling, and in order to receive this, a MUSE decoder for returning the band-compressed signal to the original wide band signal is required. . Examples of this MUSE decoder include, for example, "MUSE Decoder", NEC Technical Report, Vol. 43N
o. There is known a structure as described in 4.1990.

FIG. 8 shows a conventional example of this MUSE decoder, and its operation will be described below. The MUSE signal input from the input terminal 1 is guided to the input processing means 3 through the low pass filter (LPF) 2. In the input processing means 3, as an input processing, an analog-digital converter (ADC) 4 converts it into a digital signal, and a non-linear de-emphasis processing means 5 performs non-linear de-emphasis processing. This input-processed video signal is guided to the inter-frame interpolation processing means 7, the field interpolation processing means 11, the low-frequency replacement processing means 14 and the motion detection processing means 15 in the luminance signal processing means 6.

Regarding the luminance signal of the input processed video signal, the luminance signal processing means 6 performs the still image processing by the inter-frame interpolation processing means 7 by inter-frame interpolation processing. LPF) 8 performs filtering processing, sampling frequency conversion processing means 9 performs sampling frequency conversion processing, and inter-field interpolation processing means 10
Then, the inter-field interpolation processing is performed, and as the moving image processing, the field interpolation processing means 11 performs the field interpolation processing and the sampling frequency conversion processing means 12 performs the sampling frequency conversion processing. The luminance signal subjected to the moving image processing and the luminance signal subjected to the still image processing are guided to the adaptive mixing processing means 13.

On the other hand, the motion detection processing means 15 generates a video from the signal of the current field from the input processing means 3 and the signal of one frame before and the signal of two frames before which are introduced from the interframe interpolation processing means 7. A moving part is detected to obtain a motion amount, and a motion signal corresponding to the motion amount is guided to the adaptive mixing processing means 13 in the luminance signal processing means 6 and the adaptive mixing processing means 20 for color signal processing.

The adaptive mixing processing means 13 adaptively mixes the luminance signal subjected to the moving image processing and the luminance signal subjected to the still image processing according to the motion signal. In the luminance signal in which the still image and the moving image are adaptively mixed in this way, the low-frequency substituting processing unit 14 substitutes the low-frequency component thereof with the low-frequency component of the luminance signal which is currently arrived and is introduced from the input processing unit 3. To be After that, the replaced luminance signal of the low frequency component is
It is guided to the inverse matrix processing means 23 of the output processing means 22.

Of the input video signals, the color signals are subjected to interframe interpolation processing at the same time as the luminance signals in the interframe interpolation processing means 7 as still image processing,
After that, the time extension processing means 16 extends the time four times,
Inter-field interpolation processing means 17 performs inter-field interpolation processing. Further, as moving image processing, field interpolation processing means 11 performs field interpolation processing at the same time as a luminance signal, and then time expansion processing means 18 expands the time four times, and field interpolation processing means 19 performs field expansion. Interpolation processing is performed. The moving image processed color signal and the still image processed color signal are guided to the adaptive mixing processing means 20.

The adaptive mixing processing means 20 detects a still image processed color signal and a moving image processed color signal as motion detection processing means.
The signals are adaptively mixed in accordance with the motion signal derived from 15, and guided to the line-sequential decoding processing means 21. Line-sequential decoding processing means
Reference numeral 21 is a line-sequential decoding of a line-sequential color difference signal (that is, a color-difference signal obtained by alternately time-division-multiplexing the color-difference signals R-Y and B-Y for each scanning line) which is guided as a color signal As processing, the color difference signals are separated into two color difference signals, line interpolation is performed on each color difference signal, and the original color difference signals R-Y, B-
Y is obtained and led to the inverse matrix processing means 23 of the output processing means 22.

In the output processing means 22, the luminance and color difference signals guided by the inverse matrix processing means 23 are converted into R
The signal is converted into a GB signal, the gamma correction processing means 24 performs gamma correction, the digital-analog conversion (DAC) 25 converts the signal into an analog signal, and the signal is output. Then, these analog RGB signals are output from the output terminals 27, 28 and 29 via the low pass filter (LPF) 26.

In the MUSE system, the color signal is time-compressed to 1/4 of the luminance signal to make the data rates of the color signal and the luminance signal the same, and the luminance signal and the color signal are time-division multiplexed and transmitted. ing. For this reason, in the MUSE decoder, the inter-frame interpolation processing means 7 and the field interpolation processing means 11 of the luminance signal processing means 6 respectively perform interpolation processing, and then the color signal is separated from the luminance signal and time The expansion processing means 16 and 18 are led to perform color signal processing.

At this time, the noise removal (noise reduction) of the signal is performed by the same means for the luminance signal and the color signal in the inter-frame interpolation processing means 7 before performing the time expansion processing for the color signal. That is, the noise reduction is performed by detecting the difference between the field of the video signal that has arrived now and the field of the video signal two frames before as noise and subtracting it from the video signal of the current field.

[0013]

As described above, in the conventional MUSE decoder, the noise reduction for the luminance signal and the chrominance signal is performed by the interframe interpolation processing means 7 before performing the time extension processing for the chrominance signal. It was done by the same means.

Therefore, for the color signal, noise reduction is performed in the MUSE signal format that is band-compressed by using subsampling, that is, as the signal in which the high frequency component is folded back to the low frequency region by the subsampling. It was

However, when an attempt is made to perform noise reduction in the interframe interpolation processing means 7 with the MUSE signal format as it is, the correlation of the signals in the signal format including the aliasing component due to the band compression in the MUSE system at that time. However, there is a problem in that it is difficult to perform noise reduction using the accurate correlation because the correlation is not detected and the accurate correlation is not detected.

Since the luminance signal represents only the brightness,
The color difference signal as a color signal has only a value in the positive direction, whereas the color difference signal as a color signal has a value in the positive direction to represent red and a negative value to represent cyan as a complementary color. Therefore, when the luminance signal and the chrominance signal (color difference signal) have the same bit width, the dynamic range of the chrominance signal is substantially half that of the luminance signal. .

Therefore, if the noise reduction for the luminance signal and the chrominance signal is performed by the same means in the interframe interpolation processing means 7, the noise reduction for the chrominance signal is inevitably inferior to the noise reduction for the luminance signal. Another problem is that the S / N of the color difference signal is 3 dB worse than that of the luminance signal.

Further, in this way, when the noise of the color signal is not sufficiently removed by the interframe interpolation processing means 7, when the color signal is time-expanded four times in the time expansion processing means 16, 18 in the subsequent stage, There is a problem that noise that cannot be completely removed is expanded four times in time, its frequency band becomes ¼, and it is converted into noise in a low-frequency region that is easily noticeable.

Therefore, an object of the present invention is to solve the above-mentioned conventional problems, to perform noise reduction utilizing accurate correlation with a color signal, and further to perform noise reduction exclusively for color signals. The purpose of the present invention is to provide a MUSE decoder with a color noise reducer that can effectively reduce noise that has been converted into a low-frequency region that is easily noticeable by time extension.

[0020]

In order to achieve the above object, according to the present invention, adaptive mixing means for mixing a still image processed color difference signal and a moving image processed color difference signal in accordance with a motion signal, Correlation between the two frames is provided between the color signal obtained by mixing and the line-sequential decoding means for decoding the color-difference signal that has been line-sequentially into two color-difference signals, or at the subsequent stage of the line-sequential decoding means. Is used to provide a color noise reducer that removes noise in the color difference signal.

[0021]

The color noise reducer performs the inter-frame / inter-field interpolating process as the still image process and the field interpolating process as the moving image process so that the aliasing component due to the band compression (that is, the aliasing component in the MUSE signal format). ) Is removed and the time extension processing is performed, noise reduction is performed on the time-extended color signal.

As a result, when noise reduction is performed, the correlation of the color signals from which the aliasing component due to band compression is removed is detected, so that more accurate correlation can be detected and the performance of noise reduction is improved. Can be achieved. Further, since the noise of the color difference signal which has been converted into the low frequency region due to the time extension and which is easily noticeable can be removed, a high-quality video signal with good S / N can be obtained.

[0023]

FIG. 1 shows a first embodiment of the present invention. In FIG. 1, reference numeral 30 is a color noise reducer, and other components are the same as those in the conventional example of FIG. Color signal processing different from the conventional example will be described below.

As a still image process, the color signal is interframe-interpolated by the interframe interpolation processing means 7, quadruple time-expanded by the time expansion processing means 16, and further interfield interpolated by the interfield interpolation processing means 17. 2 by interpolating
The sub-sampling degree removes the aliased components. Further, the color signal is subjected to moving image processing by removing the aliasing component generated by the subsampling by the field interpolation means 11, expanded by four times by the time expansion processing means 18, and field-interpolated by the field interpolation processing means 19. Interpolated.

Thereafter, the color signal subjected to the moving image processing and the color signal subjected to the still image processing are adaptively mixed in the adaptive mixing processing means 20 based on the motion signal guided from the motion detection processing means 15. The color signal in which the still image and the moving image are mixed is guided to the color noise reducer 30.

The color noise reducer 30 receives the line-sequential color difference signals (that is, the color difference signals R-Y, BY) as color signals.
Noise reduction is performed by utilizing the fact that the color difference signals obtained by alternately time-division-multiplexing for each scanning line have a correlation between two frames. The noise-reduced line-sequential color difference signal is guided to the line-sequential decoding processing means 21. After that, the same processing as in the conventional example of FIG. 8 is performed, and the original wide band high-definition video signal is output from the output terminals 27, 28 and 29.

As described above, since the noise reduction is performed after the aliasing component is removed from the color signal converted into the MUSE system and including the aliasing component, the correlation of the color signals can be accurately detected. It is possible to improve the performance of noise reduction.

Next, the color noise reducer 30 shown in FIG.
2 shows an example of a specific configuration of the above. In FIG. 2, 201
Is a line-sequential color difference signal input terminal, 202 is a first multiplier that multiplies a multiplier (1-K), 203 is an adder, 204 is a second multiplier that multiplies a multiplier K, and 205 is output from the adder 203 Two-frame delay means for delaying the line-sequential color difference signal by two frames, 206
Is a terminal for outputting the addition output of the adder 203. Less than,
The operation of the noise reduce process for the line-sequential color difference signal will be described.

The line-sequential color difference signal is input to the terminal 201 and guided to the first multiplier 202. The multiplier 202 is a multiplier (1-K)
Is multiplied by the line-sequential color difference signal. The line-sequential color difference signal multiplied by the multiplier (1-K) is guided to the adder 203. The adder 203 adds the outputs of the first multiplier 202 and the second multiplier 204 and outputs the result to the terminal 206. 2 frame delay means 205 is adder 203
Output signal is delayed by 2 frames and output. Second multiplier
Reference numeral 204 denotes a signal obtained by delaying two frames by the two-frame delay means 205 by a multiplier K and adding the result of the multiplication to an adder 203.
Output to.

The data of the line-sequential color difference signal is 1 in 2 frames.
There are correlations at intervals of these two frames. Where K
The range of the value of 0 is 0 <K <1. For example, if 0.5 is given as the value of K, S / N can be improved by 3 dB and a noise-reduced line-sequential color difference signal can be obtained at the terminal 206. be able to.

Therefore, according to this configuration example, noise reduction can be performed on the line-sequential color difference signal. In addition,
The noise reducer shown in FIG. 2 is an example, and the present invention is not limited to this configuration.

As described above, according to this embodiment, the MUS
Since the noise reduction process can be performed after the signal processing for removing the aliasing component is performed on the color signal including the aliasing component in the frequency by the E method, the correlation between the color signals can be accurately detected. It is possible to improve the performance of noise reduce and obtain a color signal with a good S / N.

Next, a second embodiment of the present invention is shown in FIG. The present embodiment differs from the embodiment of FIG. 1 in that an adaptive color noise reducer 31 is used instead of the color noise reducer 30 and noise reduction is performed from the motion detection processing means 15 in accordance with a motion signal. Yes, the others are the same as the embodiment of FIG. The operation will be described below with respect to points different from the embodiment of FIG.

The adaptive color noise reducer 31 performs noise reduction by utilizing the fact that the line-sequential color difference signals introduced as color signals have a correlation between two frames. At this time, the adaptive color noise reducer 31 uses a motion signal, which is introduced from the motion detection processing means 15 and indicates in some stages whether or not there is motion, in an area where the motion of the video is large in accordance with the size of the motion. The noise reduction amount is reduced, and the noise reduction amount is increased in an area where the motion of the image is small to adaptively reduce the noise of the line-sequential color difference signal. Then, the line-sequential color difference signal adaptively subjected to noise reduction is guided to the line-sequential decoding processing means 21. After that, the same processing as in the conventional example of FIG. 8 is performed, and the original wide band high-definition video signal is output from the output terminals 27, 28 and 29.

Next, FIG. 4 shows a specific structural example of the adaptive color noise reducer 31 shown in FIG. In FIG. 4, 401 is a terminal for inputting a motion signal indicating the motion area of the image guided from the motion area detecting means 15, 402 is a multiplier K, and the value of K is varied by the motion signal guided from the terminal 401. The third multiplier 403 is a fourth multiplier that multiplies the multiplier (1-K) and changes the value of K according to the motion signal guided from the terminal 401. Others are the same as the configuration example of FIG. . The operation of noise reduction processing of the line-sequential color difference signal will be described below.

The line-sequential color difference signal is input to the terminal 201 and guided to the third multiplier 402. The multiplier 402 changes the value of K according to the motion signal guided from the terminal 401, increases the multiplier K in a region where there is motion, decreases the multiplier K in a region where there is no motion, and linearly changes the multiplier K. Multiply the color difference signal. The line-sequential color difference signal multiplied by the multiplier K according to the amount of motion of the image is guided to the adder 203. The adder 203 is the third multiplier 402
And the output of the fourth multiplier 403 are added and output to the terminal 206. The frame delay means 205 delays the output signal of the adder 203 by 2 frames and outputs it. The fourth multiplier 403 changes the value of K according to the motion signal guided from the terminal 401, reduces the multiplier (1-K) in the area where there is motion, and the multiplier (1-K) in the area where there is no motion. ) Is increased to multiply the signal delayed by two frames by the two-frame delay unit 205 by the multiplier (1-
K) and outputs the multiplication result to the adder 203.

The data of the line-sequential color difference signal is 1 in 2 frames.
There are correlations at intervals of these two frames. Where K
The range of values of 0 is 0 <K <1, and for example, the value of K is 0.9 in the case of a completely moving region and 0.1 in the case of a still image without motion. Depending on the amount of movement, these values may be divided into several stages to digitally control the value of K, or the range of 0 <K <1 may be analogized according to the amount of movement. You may control with. Therefore, according to the present configuration example, it is possible to adaptively perform noise reduction on the line-sequential color difference signal according to the amount of motion of the image.

In this configuration example, the value of K is 0 <
Since it can be freely changed within the range of K <1, the value of K is set to, for example, _KRY corresponding to the _RY signal in accordance with the switching of the RY signal / BY signal in the line-sequential color difference signal.
It is also possible to control so as to switch alternately between K _BY and K _BY corresponding to the _BY signal. In such a case, in accordance with the reaction of the human eye to the color, the line sequential color difference signal It is possible to perform appropriate noise reduction on the R-Y signal and the B-Y signal in FIG.

Next, FIG. 5 shows another concrete example of the configuration of the adaptive color noise reducer 31 shown in FIG. 5, 501 is a first subtractor, 503 is a second subtractor, 502 is a fifth multiplier, and the others are the same as the configuration example of FIG.
The operation of noise reduction processing of the line-sequential color difference signal will be described below.

The subtractor 501 subtracts the output signal of the fifth multiplier 502 from the line-sequential color difference signal guided from the input terminal 201, and outputs it to the output terminal 206. The 2-frame delay means 205 delays the signal guided from the first subtractor 501 by 2 frames and outputs it to the second subtractor 503. The second subtractor 503 is
The two-frame front line sequential color difference signal guided from the two-frame delay means 205 is subtracted from the line sequential color difference signal guided from the terminal 201, and output to the fifth multiplier 502. Fifth multiplier 5
02 changes the value of K in accordance with a motion signal indicating the motion region of the image guided from the terminal 401, decreases the multiplier K in a region where there is motion, and increases the multiplier K in a region where there is no motion. The signal derived from the second subtractor 503 is multiplied by the multiplier K and output to the first subtractor 501.

Here, the value of K is in the range of 0 <K ≦ 0.5, and a value close to 0 when the motion amount is large and a value close to 0.5 when the still image is not moving. Control to take. Therefore, according to this configuration example as well, noise reduction can be adaptively performed on the line-sequential color difference signals in accordance with the amount of motion of the image.

In this configuration example as well, the value of K is 0 <
Since it can be freely changed within the range of K ≦ 0.5, the value of K is set to, for example, the value corresponding to the RY signal in accordance with the switching of the RY signal / BY signal in the line-sequential color difference signal.
It is also possible to control so as to switch alternately between _RY and K _BY corresponding to the _BY signal, and in such a case, the line-sequential color difference is adjusted according to the reaction of the human eye to the color. Appropriate noise reduction can be performed on the RY signal and the BY signal in the signal.

The configuration example of the adaptive color noise reducer 31 in FIG. 3 has been described above. However, the configuration example is not limited to that shown in FIG. 4 or FIG. Anything that can reduce noise is acceptable.

As described above, according to this embodiment, the MUS
By the E method, it is possible to adaptively perform noise reduction processing after performing signal processing for removing the aliasing component on the color signal including the aliasing component in the frequency. The noise reduction performance can be improved by detecting the noise, and noise reduction can be performed according to the still image area and the moving image area of the image, and the S / N is good, and the color of the image in the moving image area is less blurred. It becomes possible to obtain a signal.

Further, according to this embodiment, the value of K of the multiplier in the adaptive color noise reducer 31 is set to, for example, the value of K of the adaptive color noise reducer 31 in accordance with the switching of the RY signal / BY signal in the line-sequential color difference signal. When control is performed so as to alternately switch between K _RY corresponding to the _RY signal and K _BY corresponding to the _BY signal, line sequential is performed according to the reaction of the human eye to color. Appropriate noise reduction can be performed on the RY signal and the BY signal in the color difference signal.

Next, FIG. 6 shows a third embodiment of the present invention. The present embodiment differs from the embodiment of FIG. 3 in that the same signal as the motion signal used in the adaptive mixing processing means 20 is delayed by a predetermined time and used in the adaptive color noise reducer 31. Others are the same as the embodiment of FIG. The operation will be described below with respect to points different from the embodiment of FIG.

In the adaptive color noise reducer 31, the same signal as the motion signal guided to the adaptive mixing processing means 20 is supplied to the delay means 32 as the motion signal indicating the area where the image moves.
Then, the process is delayed by a predetermined time, that is, the processing time required from the adaptive mixing processing to the noise reduction processing. As a result, the motion signal indicating the motion of the image is correctly guided in time, so that noise reduction according to the motion of the image can be performed.

As described above, also in the present embodiment, similarly to the embodiment of FIG. 3, after the color signal including the aliasing component in the frequency is subjected to the signal processing for removing the aliasing component by the MUSE method. Since noise reduction processing can be applied adaptively, the noise reduction performance can be improved by accurately detecting the correlation of color signals, and noise reduction depending on the still image area and moving image area of the image. Therefore, it is possible to obtain a color signal with good S / N and less blurring of the image in the moving image area. Further, according to the present embodiment, the motion detection processing means
Since the motion signal indicating the moving part of the video output from 15 can be shared by the adaptive mixing processing means 20 and the adaptive color noise reducer 31, the motion detection processing means 15 can easily detect the motion of the video. become.

Next, a fourth embodiment of the present invention is shown in FIG. The present embodiment is different from the embodiment of FIG. 3 in that the adaptive color noise reducer 31 arranged before the line-sequential decoding processing means 21 is replaced with the adaptive color noise reducer 33 after the line-sequential decoding processing means 21. And two color difference signals R- obtained by line-sequential decoding in the line-sequential decoding processing means 21 by the adaptive color noise reducer 33.
The noise reduction process is applied to Y and BY, and the other points are the same as those in the embodiment of FIG. Below, FIG.
The operation will be described with respect to points different from the embodiment of FIG. The operation up to the line-sequential decoding processing means 21 is the same as in the conventional example of FIG.

Two color difference signals RY and BY obtained by line-sequential decoding in the line-sequential decoding processing means 21.
Is guided to the adaptive color noise reducer 33. The adaptive color noise reducer 33 is configured by using, for example, the two configuration examples shown in FIG. 4 for the RY signal and the BY signal, for example. By the motion signal indicating the presence / absence of motion, the noise reduction amount is increased in the still image part having no motion, and the noise reduction amount is reduced in the motion image part having the motion, and the color difference signals RY and BY are respectively supplied. Apply noise reduce processing. This noise-reduction-processed color difference signal R-
Y and BY are guided to the output processing means 22.

As described above, also in the present embodiment, similarly to the embodiment of FIG. 3, after the color signal containing the aliasing component in the frequency is subjected to the signal processing for removing the aliasing component by the MUSE method. Since noise reduction processing can be applied adaptively, the noise reduction performance can be improved by accurately detecting the correlation of color signals, and noise reduction depending on the still image area and moving image area of the image. Therefore, it is possible to obtain a color signal with good S / N and less blurring of the image in the moving image area.

Also in this embodiment, as in the embodiment of FIG. 6, the same signal as the motion signal used in the adaptive mixing processing means 20 is delayed by a predetermined time and used in the adaptive color noise reducer 33. You can Also in this embodiment, the color noise reducer as shown in FIG. 2 may be used as in the embodiment of FIG. 1 for the purpose of simplifying the color noise reducer.

[0053]

According to the present invention, for a line-sequential color difference signal having a folding component in the frequency band of the signal by band compression by the MUSE method, interframe interpolation and interfield interpolation are performed in the still image area. Insertion processing,
In the moving image area, noise reduction processing can be performed after performing field interpolation processing to remove aliasing components and returning to the original wideband signal. Therefore, it is possible to accurately detect the correlation of the color signals, perform noise reduction using the accurate correlation, and improve the noise reduction performance. Moreover, noise reduction can be performed only for the color signal, and further, noise reduction processing can be effectively performed on noise that is more noticeable due to conversion to the low frequency region by time extension. Therefore, it is possible to obtain a color signal having a good S / N and less blurring of the image in the moving image area.

[Brief description of drawings]

FIG. 1 is a block diagram showing a first embodiment of the present invention.

FIG. 2 is a block diagram showing a specific configuration example of the color noise reducer in FIG.

FIG. 3 is a block diagram showing a second embodiment of the present invention.

FIG. 4 is a block diagram showing a specific configuration example of the color noise reducer in FIG.

5 is a block diagram showing another specific example of the configuration of the color noise reducer shown in FIG.

FIG. 6 is a block diagram showing a third embodiment of the present invention.

FIG. 7 is a block diagram showing a fourth embodiment of the present invention.

FIG. 8 is a block diagram showing a conventional example of a MUSE decoder.

[Explanation of symbols]

1 ... MUSE signal input terminal, 2 ... low-pass filter, 3
... Input processing means, 4 ... Analog-digital converter, 5
... Non-linear de-emphasis processing means, 6 ... Luminance signal processing means, 7 ... Inter-frame interpolation processing means, 8 ... Still image low-pass filter, 9 ... Sampling frequency conversion processing means, 10 ... Inter-field interpolation processing means, 11 ... Field interpolation processing means, 12 ... sampling frequency conversion processing means,
13 ... Adaptive mixing processing means, 14 ... Low range replacement processing means, 15
... Motion detection processing means, 16 ... Time expansion processing means, 17 ... Inter-field interpolation processing means, 18 ... Time expansion processing means, 19 ... Field interpolation processing means, 20 ... Adaptive mixing processing means, 21 ...
Line-sequential decoding processing means, 22 ... Output processing means, 23 ... Inverse matrix processing means, 24 ... Gamma correction processing means, 25 ... Digital-analog conversion means, 26 ... Low-pass filter, 27
... R signal output terminal, 28 ... G signal output terminal, 29 ... B signal output terminal, 30 ... color noise reducer, 31 ... adaptive color noise reducer, 32 ... delaying means, 33 ... adaptive color noise reducer, 201 ... wire Sequential color difference signal input terminal, 202 ... First multiplier, 203 ... Adder, 204 ... Second multiplier, 205 ... Two frame delay means, 206 ... Video signal output terminal, 401 ... Motion signal input terminal, 402 ... Third multiplier, 403 ... Fourth multiplier, 501
... first subtractor, 502 ... fifth multiplier, 503 ... second subtractor.

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Masato Sugiyama, 292 Yoshida-cho, Totsuka-ku, Yokohama-shi, Kanagawa Inside Visual Media Research Laboratory, Hitachi, Ltd. (72) Inventor Kentaro Teranishi 292 Yoshida-cho, Totsuka-ku, Yokohama, Kanagawa Hitachi, Ltd., Visual Media Laboratory (72) Inventor, Hatsuji Kimura, 292 Yoshida-cho, Totsuka-ku, Yokohama, Kanagawa Prefecture 6-chome, Hitachi, Ltd. (72) Inventor, Yuichi Ninomiya 1-1-10 Kinuta, Setagaya-ku, Tokyo, Japan Broadcasting Corporation Research Institute

Claims (3)

[Claims] 1. A MUSE (Multiple Sub-Nyquist Sampling) obtained by band-compressing a wideband high-definition signal.
In a MUSE decoder that decodes an Encoding) signal into an original wideband high-definition signal, a wideband still image is obtained by interpolating a luminance signal of at least four fields among the pre-digitized MUSE signals between fields. First static area processing means for decoding into an image luminance signal, and by interpolating the luminance signal of the digitized MUSE signal for one field, the original broadband moving image luminance signal is obtained. First motion area processing means for decoding,
Of the digitized MUSE signal, a motion amount detecting means for detecting a motion amount between frames or fields of a luminance signal and outputting a motion signal according to the motion amount, and the stationary obtained by decoding A first mixing unit that mixes the image brightness signal and the moving image brightness signal according to the motion signal output from the motion amount detection unit, and a brightness signal obtained by mixing the first brightness mixing unit and the first brightness mixing unit. Of these, the low-frequency component is replaced with the low-frequency component of the luminance signal of the digitized MUSE signal, or the first low-frequency component is replaced.
Low-pass substituting means for mixing the low-frequency component of the luminance signal of the digitized MUSE signal with the low-frequency component of the luminance signal obtained by mixing by the mixing means of a certain ratio; Second still area processing means for decoding a line-sequential color difference signal among the converted MUSE signals into inter-field interpolation using at least four fields to thereby obtain a wide-band still image color difference signal; Second moving area processing means for decoding the line-sequentialized color difference signal in the digitized MUSE signal into the original broadband moving image color difference signal by interpolating the field by one field; Second mixing means for mixing the still image color difference signal and the moving image color difference signal obtained by decoding in accordance with the motion signal output from the motion amount detecting means. And line-sequential decoding means for decoding the color-difference signal obtained by mixing by the second mixing means into two color-difference signals from the line-sequential color-difference signal, and the second mixing means. A color noise reducer for removing noise of the color difference signal by utilizing the correlation between two frames is provided between the line-sequential decoding means and a stage subsequent to the line-sequential decoding means. M
USE decoder. 2. The MUSE decoder with color noise reducer according to claim 1, wherein the color noise reducer removes noise of the color difference signal according to the motion signal output from the motion amount detecting means. A MUSE decoder with a color noise reducer. 3. The MUSE decoder with color noise reducer according to claim 2, wherein the motion signal used in the color noise reducer among the motion signals output from the motion amount detecting means is the second mixing means. A MUSE decoder with a color noise reducer, which is the same signal as a motion signal used in.