JP2675354B2 - Decoding device for MUSE signal - Google Patents

Description

The present invention relates to a device for decoding a MUSE signal that is band-compressed and transmitted in a high-definition receiver, and particularly, it is easy to improve the image quality and is an NTSC television. The present invention relates to a decoding device suitable for compatibility with signal processing.

[Conventional technology]

There is a high-definition system as a television system being developed as a next-generation television system. Since this system requires a transmission band about five times that of the current NTSC system, it is necessary to perform band compression to reduce the transmission band for actual broadcasting. As one concrete method of this transmission method, Nikkei Electronics "High-definition television, difficult global standard", 1987.8.10 (n
o.427), and the MUSE method described on pages 97 to 112. As described, this method is a method of compressing data to 1/4 and transmitting the data by offset sampling the image between four fields on the transmitting side. Therefore, on the receiving side, the MUSE signal in which the data is compressed to 1/4 must be restored to the original wideband TV signal by using a frame memory or the like. This band compressed
MU restores the MUSE signal to the original wideband television signal
It is an SE decoding device.

As described in the above-mentioned document, the format of the MUSE signal is such that the chrominance signal is time-compressed to 1/4 and is line-sequentially time-division multiplexed in the blanking period of the luminance signal. Therefore, in the decoder device on the receiving side, the luminance signal and the color signal are continuously processed in a time-division manner, and the same frame memory, still region interpolation, motion region interpolation circuit, and the like are used for the luminance signal and the color signal. It was

[Problems to be solved by the invention]

In the above-mentioned conventional technique, since the luminance signal and the color signal are time-division-multiplexed, the characteristics for decoding the luminance signal and the characteristics for decoding the color signal are not considered so that they can be set arbitrarily. There is a problem that it is difficult to optimize the decoding characteristics of each signal according to the conditions of the luminance signal and the color signal.

In addition, in the current NTSC system, the luminance signal and the color signal are frequency-multiplexed, and time-division processing cannot be performed like the MUSE signal, so in the conventional technology, the current NTSC system is digitally processed to improve image quality. That is, there is a problem that it is difficult to combine the IDTV function.

An object of the present invention is to optimize the decoding characteristics of the luminance signal and the chrominance signal in the MUSE decoding device so that the image quality can be further improved.

Another object of the present invention is to provide a MUSE decoding device in which the IDTV function for improving the image quality of NTSC signals can be easily combined.

[Means for solving the problem]

The above-mentioned object is, in the MUSE decoding device, after the analog-digital conversion of the MUSE signal, a separation circuit for separating the time-division multiplexed luminance signal and chrominance signal is provided, and a frame memory for luminance signal and chrominance signal. And each interpolation circuit separately provided, and for a color signal that is time-compressed to 1/4, it is time-expanded in the latter stage of the separation circuit and then decoded by the color signal frame memory and each interpolation circuit, etc. Achieved by

[Action]

In the above means, the luminance signal separated by the separation circuit can be subjected to the same decoding process as the conventional MUSE decoding device to set the optimum characteristics for the luminance signal, and separate circuits for the color signal after separation must be provided separately. This makes it possible to set the optimum decoding characteristic for the color signal. Further, by performing the color signal decoding after the time extension, the operation speed of the color signal decoding unit can be reduced, and the margin of the timing setting of the sub-sample clock required for each decoding is significantly increased. Timing malfunction can be suppressed and the image quality can be improved.

Also, because the luminance signal decoding circuit and the chrominance signal decoding circuit are separately provided, it is possible to use both the luminance signal and chrominance signal processing circuits required for NTSC IDTV processing. Becomes

〔Example〕

An embodiment of the present invention will be described below with reference to the circuit block diagram of the MUSE decoding device shown in FIG.

In FIG. 1, 1 is an input terminal for the MUSE signal, 2-4 are output terminals for the R, G, B signals that have been restored to the original wideband television signals by the decoding device, and 5 and 6 are HD, VD signals. An output terminal, 7 is an analog-digital converter (hereinafter referred to as an A / D converter) for converting an analog MUSE signal to digital,
Reference numeral 8 is a de-emphasis circuit for de-emphasising the digitally converted MUSE signal, 9 is a synchronization processing circuit, and 10 is a separation circuit for separating the time-division multiplexed luminance signal (Y) and chrominance signal (C). 11 to 19 are luminance signal decoding units, 20 to 30 are chrominance signal decoding units, 31 is a luminance signal decoding unit, and the luminance signal restored to the original wideband signal and the chrominance signal decoding unit are the original wideband signals. Color difference signal (for example, RY
An inverse matrix circuit for converting a signal and a BY signal) into original signals of R, G, B, and 32 is a digital-analog converter (hereinafter referred to as a D / A converter). In the luminance signal decoding unit, 11 is a luminance signal inverse gamma correction circuit (hereinafter, Cγ
^-1 circuit), 12 is a luminance signal noise reduce circuit (hereinafter referred to as YNR circuit) 13, 14 is a frame memory (hereinafter referred to as Y memory), and 15 is a frame interpolation process for still images. Circuit (hereinafter referred to as Y still image processing circuit), 16 is a moving image frame interpolation processing circuit (hereinafter referred to as Y moving image processing circuit), and 17 is a still image inter-field interpolation processing circuit (hereinafter, referred to as Y moving image processing circuit). Y field processing circuit), 18 is a mix circuit for mixing a still image processed luminance signal and a moving image processed luminance signal (hereinafter referred to as YMIX circuit), 19 is a luminance signal motion amount detection circuit. (Hereinafter, referred to as Y motion detection circuit.)
It is. In the color signal decoding unit, 20 is a time decompression circuit (hereinafter referred to as C time decompression circuit) for returning a 1/4 time-compressed color signal to the original time axis. CNR circuits), 22 and 23 are frame memories (hereinafter referred to as C memories), 24 is a still image interframe interpolation processing circuit (hereinafter referred to as C still image processing circuit), 25
Is a moving image frame interpolation processing circuit (hereinafter referred to as a moving image processing circuit), 26 is a still image inter-field interpolation processing circuit (hereinafter referred to as C field processing circuit), and 27 is a mix circuit (hereinafter referred to as C field processing circuit). 28 is a color signal motion amount detection circuit (hereinafter, referred to as C motion detection circuit), and 29 is a line-sequential color difference color signal (for example, RY and BY signals). A color decoding circuit (hereinafter, referred to as a C decoding circuit) 30 for returning to the simultaneous color difference signal is a color signal inverse gamma correction circuit (hereinafter, referred to as a C signal decoding circuit).
It is referred to as a Cγ ^-1 circuit. ).

 The operation of each circuit will be briefly described below.

First, the operation of the luminance signal decoding unit will be described. The current luminance signal separated in the separation circuit 10 is Yγ
^The transmission inverse gamma correction is performed by the ^-1 circuit 11, and the result is guided to the YNR circuit 12. The Y memories 13 and 14 are circuits for delaying each by one frame, and the luminance signals of one frame before and two frames before are output to the outputs of the Y memories 13 and 14, respectively. In the YNR circuit 12,
Noise reduction is performed by utilizing the correlation between the current luminance signal and the signal two frames before from the Y memory 14. Y
In the still image processing circuit 15, the noise reduced current luminance signal and the luminance signal of one frame before are introduced, subjected to horizontal and vertical filter processing and three-dimensional processing by inter-frame interpolation, and banded by offset sampling between fields and frames. For the compressed muse signal, the band compression amount by the inter-frame offset sampling is restored. The Y field processing circuit 17 restores the band compression amount due to the inter-field offset sampling. Therefore, the original broadband luminance signal is obtained at the output of the Y field processing circuit 17. On the other hand, in the Y moving image processing circuit 16,
Since the moving image moves in the time axis direction, the intra-field processing by the horizontal and vertical two-dimensional processing is performed only with the current luminance signal from the YNR circuit 12. In the YMIX circuit 18, the signal processed by the still image and the moving image as described above is supplied to the Y motion detection circuit.
By mixing according to the amount of movement between one frame or two frames detected in 19, by interpolating the signal processed by the moving image in the moving image area and the signal processed by the still image in the still image area, , The output of the YMIX circuit 18 can obtain a luminance signal that has been subjected to motion adaptation processing without unnaturalness.

 Next, the operation of the color signal decoding unit will be described.

The chrominance signal separated in the separation circuit 10 is time-expanded four times in the C time expansion circuit 20, and then the current color signal noise-reduced by the CNR 21 and the C memories 22 and 23 in the same manner as the luminance signal. Pre-frame color signal is converted to C still image processing circuit 24
It is led to. By the C still image processing circuit 24 and the C field processing circuit 25, the same three-dimensional processing as the luminance signal is performed, and the original wideband color signal is restored. C video processing circuit 26
The current color signal is introduced into the signal, and the same two-dimensional processing as the luminance signal is performed. In the CMIX circuit 27, the color signals processed as a still image and a moving image as described above are mixed according to the amount of motion detected by the C motion detection circuit 28, and a color signal subjected to motion adaptive processing without unnaturalness is obtained. can get. The color signal subjected to the motion adaptive processing is, for example, RY and B- in the C decoding circuit 29.
The Y line-sequential color difference signal is returned to the simultaneous signal, and Cγ ^-1
Transmission inverse gamma correction is applied to the color signals synchronized by the circuit 30.

The signals returned to the original wide-band luminance signal Y and the simultaneous color difference signals RY and BY by the motion adaptive processing as described above are inversely converted into the original signals R, G and B by the inverse matrix circuit 31. ,
It is converted into an analog signal by a digital-analog conversion circuit 32 (hereinafter referred to as a D / A converter), and guided to a monitor from terminals 2-4.

A feature of the present invention in the above circuit configuration is that a separation circuit 10 is provided, and a color signal decoding unit made of 20 to 30 is provided separately from a luminance signal decoding unit made of 11 to 19 in a subsequent stage,
In addition, the C time expansion circuit is provided in front of the color signal decoding units 20 to 30.
20 is provided.

As described above, by providing the color signal decoding units 20 to 30 separately from the luminance signal decoding units 11 to 19, it is possible to arbitrarily set the noise reduction characteristics of the color signal regardless of YNR12. FIG. 2 is a diagram showing an example of the noise reduce circuit. The 12 surrounded by the broken line is 3 for the YNR circuit.
4 is a subtractor, 35 is a non-linear coefficient unit, and 36 is an adder, which is a non-linear coefficient () for the difference signal between the current signal from the terminal 33 and the two-frame-preceding signal from the frame memories 13 and 14. For example, it is multiplied by α) and added to the current signal from the terminal 33.
In this case, the coefficient unit 35 has a value of, for example, 1/2 when the difference signal level from the subtractor 34 is small, and the coefficient α decreases when the difference signal level becomes a certain value or more, and becomes zero when the difference signal level is large. Become. Thereby, for example, the high frequency components of the moving image portion are prevented from being deteriorated by the noise reduce circuits 12 and 21.
However, the color signal has a narrower band than the luminance signal, and the deterioration of the band is less noticeable especially in moving images, and the S /
Image quality deterioration due to N deterioration is more noticeable. In the present invention, Y
Since the NR12 and the CNR21 are separately provided, it becomes possible to optimize the noise reduce characteristic for color signals for color signals.

In the present invention, the still image processing circuits 15 and 24 and the inter-field processing circuits 17 and 26 are the same as the noise reduce circuits 12 and 21.
The moving image processing circuits 16 and 26 are separately provided for the luminance signal and the color signal. Therefore, similarly to the noise reduce circuits 12 and 21, the still image processing characteristic and the moving image processing characteristic of the color signal can be set to the optimal characteristic for the color signal separately from the luminance signal.

In the present invention, the still image processing circuits 15 and 24 are used to
The inter-frame sub-sampling clock for re-sample interpolating the pre-frame signal and the inter-field sub-sampling clock for inter-field re-sampling in the inter-field processing circuits 17 and 26 for luminance signal and chrominance signal And can be supplied separately, and the phase setting of each sub-sample clock becomes easy.

FIG. 3 is a diagram for explaining the sub-sampling phase of the luminance signal. In FIG. 3, n, n + 1, n + 2, n + 3 indicate line numbers of odd fields, and m, m + 1, m + 2, m + 3 indicate line numbers of even fields. In addition, the MUSE signal has been subjected to offset sampling of 4 fields, and 3a ○
The mark indicates the sampling phase of the first field, the mark of 3b indicates the sampling phase of the second field, and the mark of 3c indicates the third phase.
The sampling phase of the field, and the symbol 3d indicates the sampling phase of the fourth field. Therefore, for example, the inter-frame sample clock is obtained by interpolating the first field data and the third field data as shown in 3a and 3c, and the sub-sample clocks of the nth line and the n + 2th line are Like 3e, the sub-sample phases of the (n + 1) th line and the (n + 3) th line are inverted to become 3f. FIG. 4 is a diagram for explaining the sub-sampling phase of the color signal. In FIG. 4, 4a to 4d indicate sampling phases of the first to fourth fields, as in 3a to 3d of FIG. However, since the color signals are line-sequentially multiplexed, the sampling phase becomes the same phase for n and n + 1, n + 2, and n + 3 for 2 lines, and for example, the phase-inverted sub-sample clocks such as 4e and 4f are every 2 lines. Alternate with.

Therefore, although not shown, the inter-frame sub-sample clocks for the luminance signal and the chrominance signal, which are guided to the Y still image processing circuit 15 and the C still image processing circuit 24 in FIG. 1, are different. Similarly, the inter-field sub-sample clocks led to the Y inter-field processing circuit 17 and the C inter-field processing circuit 26 are also different. FIG. 5 is an example of a MUSE signal in which a color signal and a luminance signal are time-division multiplexed, and as shown in 5a, 94 clocks out of 480 clocks of a sample of one line are color signals and 374 clocks. Are assigned to the luminance signal. Therefore, in the conventional example, it is necessary to switch each sample clock between the chrominance signal sub-sample clock and the luminance signal sub-sample clock by clock distribution as shown in 5b, for example. However, as is clear from FIG. 5, the switching margin of the sub-sample clock is only ± 3 clocks, which makes circuit design difficult. In addition, it has caused a malfunction. However, by using the present invention, it is sufficient to switch the sub-sampling phase for each line for the luminance signal in the color signal period of 100 clocks, and the switching margin is increased. Further, the switching of the sub-sampling phase for every two lines for the color signal can be switched in substantially the same phase period as the luminance signal as described later, and the sub-sampling frequency is also reduced. Therefore, the circuit design becomes easy.

Next, in the present invention, by providing the C time expansion circuit 20 in front of the color signal decoding units 20 to 30, sampling data 480 _H ( _H is a horizontal frequency of high-definition TV is time-compressed and multiplexed to 1/4. , 480 _H is about 16.2 MHz
Becomes ) Is frequency down to 120 _H by time extension. Therefore, the operating frequency of CNR21, C memory 22 and 23 is
It is reduced to 120 _H (about 4 MHz), and the sub-sampling frequency is also reduced, making it easier to make ICs, for example.

FIG. 6 is a diagram showing the timings of the luminance signal and the chrominance signal guided to the YNR12 and CNR21, 6a is a MUSE signal with a sampling frequency of 480 _H guided to the separation circuit 10, and 6b is guided from the separation circuit 10 to YNR12. A luminance signal, 6c is a color signal introduced to the C time expansion circuit 20. The C time expansion circuit 20 writes the color signal 6c to the memory with a write clock having a period width shown by 6d at a frequency of 480 _H and a frequency of 120 _{H, for example.}
Then, by reading the color signal from the memory with the read clock having the period width shown in 6e, the time-expanded color signal shown in 6f is output and guided to the CNR circuit 21.

By doing so, the timing of the luminance signal guided to the YNR circuit 12 and the timing of the color signal guided to the CNR circuit 21 substantially match. Therefore, the timings of the Y memories 13 and 14 and the C memories 22 and 23 to which these luminance signal and chrominance signal are introduced are almost the same, and although not shown, the memories 13, 14, 22, 23 are not shown.
The timing design of the control circuit can be facilitated, and the circuit can be shared. Further, the phase switching timing of each sub-sample clock becomes almost the same, the timing design becomes easy, the malfunction of the system can be reduced, and the image quality can be improved.

FIG. 7 shows a MUSE decoding device of the present invention, for example, NTSC.
FIG. 1 is a diagram showing an example in which an IDTV function of a television signal of a system is also used.

In FIG. 7, 37 is an input terminal for NTSC color signals,
38 and 39 are input terminals for each NTSC sync signal, 40 is terminal 37
The color signal A / D converter 41, a circuit 41 for demodulating a color signal having a subcarrier (hereinafter referred to as a C demodulation circuit), and others are the same as those in the embodiment of FIG. The solid line signal flow in FIG. 7 indicates the IDTV function signal flow, and the broken line indicates the first line.
In the signal flow of the MUSE decoding system shown in the embodiment of the figure, the signal flow not used during the IDTV function is shown.

The operation of the IDTV function will be described below. The brightness signal of NTSC system from terminal 1 is converted to 4 by A / D converter 7, for example.
_SC ( _SC is the color subcarrier frequency, 4 _SC is about 14.3MH
z. ) If it was sampled at the clock,
If the gradation is 8 bits, the memory capacity for one frame is given by the formula (1) 8bit × 910 × 5253.82Mbit …… (1). On the other hand, in the MUSE system, the memory capacity for one frame is given by formula (2) 8bit x 480 x 11254.32Mbit ...... (2), and in the brightness area only 8bit x 380 x 11253.42Mbit ...... (3) This is almost the same as the one-frame memory capacity of the system. Therefore, the luminance signal sampled by the A / D converter 7 is subjected to noise reduction by the YNR circuit 12 using the correlation with the luminance signal of the previous frame from the Y memory 13. The Y still image processing circuit performs a frame brightness comb filter function in which the current brightness signal from the YNR circuit 12 and the signal one frame before from the Y memory 13 are added. Further, the inter-field processing circuit 17 is provided with a field progress scan function that double-speeds the current signal and the signal one field before and converts it into a non-interlaced signal. These two functions are still image processing functions. On the other hand, in the Y moving image processing circuit 16, the line luminance comb-shaped filter function and the LPF function for suppressing the color subcarrier band are executed in parallel, both are double-speed converted, and the line progress scan for reading the same line signal twice is performed. Function is applied. In the moving image processing, the processing having the line luminance comb filter function and the line progress scanning function and the processing having the LPF function and the line progress scanning function are simultaneously executed, and each output is guided to the YMIX circuit 18. Similar to the MUSE decoding, the YMIX circuit 18 responds to the motion from the Y motion detection circuit 19 to, for example, when the luminance signal and the chrominance signal do not move, the still image processed signal does not move between frames. However, if there is no signal change between lines, the video processing signal with line brightness comb filter and line progress scan function is applied, and if there is also a signal change between lines, LPF and line progress scan function Motion adaptive processing is performed so that the moving image processed signal subjected to the above is output, and the luminance signal subjected to the IDTV system processing is obtained at the output of the YMIX circuit 18.

On the other hand, the color signal sampled by the A / D converter 40 is C
In the demodulation circuit 41, in the example, the RY and BY color difference signals are demodulated and guided to the CNR circuit 21. The band of the demodulated chrominance signal is narrower than that of the luminance signal.For example, LPF of about 500 KHz
After passing through, the sampling frequency can be reduced to 1/4 ( _SC ). Further, in this case, if the color signals are converted line-sequentially and guided to the CNR circuit 21, the C memory 22,
The memory capacity for 1 frame of 23 is the formula (4) It is almost the same as the formula (5) 8bit × 100 × 11250.9Mbit ・ ・ ・ (5) of the memory capacity for one frame in the MUSE system, which is a value suitable for dual use of memory. CNR
The circuit 21 is different from the YNR circuit 12 in that noise reduction using the correlation between two frames of color signals is performed. The C still image processing circuit 24 and the inter-field processing circuit 26 are provided with the same frame color comb filter function and field progress scan function as the luminance signal, and the C moving image processing circuit 25 is also a line color comb filter, LPF and line progress scan function. Then, the CMIX circuit 27 performs motion adaptation processing according to the amount of motion from the C motion detection circuit 28.

As described above, the IDTV system-processed luminance signal and the RY and BY color difference signals are guided to the inverse matrix circuit 31, and the D / A converter 32 converts the IDTV R, G, and B signals. The original signal is output.

As described above, by using this device, it is possible to make the system convenient when the MUSE decoding device is also used as the IDTV system.

Further, in the embodiment of FIG. 7, the color signals are processed in the color difference line sequence, but the memory capacity of the C memories 22 and 23 is doubled, the color difference signals are interpolated for each dot, and the data frequency is set to 2.
_{By using SC} , it is possible to process as simultaneous color difference signals.

Although not shown, the C demodulation circuit 41 is used for analog processing, and instead of the A / D converter 40 for color signals, it is for RY and B-.
By providing two A / D converters for Y, the sampling frequency of these two A / D converters can be reduced to _SC , and about 500K to reduce the sampling frequency.
The Hz LPF is also unnecessary.

〔The invention's effect〕

According to the present invention, the color signal decoding characteristic of the MUSE signal can be arbitrarily set separately from the luminance signal, and the system configuration of the decoding device can be simplified to reduce malfunction of decoding operation and improve image quality. There is an effect that can be achieved.

In addition, if you receive an NTSC TV on this device,
There is an effect that the IDTV function can be shared.

[Brief description of the drawings]

FIG. 1 is a block diagram showing an embodiment of the present invention, FIG. 2 is a block diagram showing an example of a noise reduce circuit, FIG. 3 is an explanatory diagram of a sub-sampling phase of a luminance signal in a MUSE system, and FIG. FIG. 5 is an explanatory view of a sub-sampling phase of a chrominance signal in the MUSE system, FIG. 5 is a waveform diagram showing an example of a multiplexed form of chrominance signal and luminance signal in the MUSE signal, and FIG. 6 is an embodiment of the present invention in FIG. FIG. 7 is a timing chart for explaining the timing of each signal introduced to the Y memory and the C memory, and FIG. 7 is a block diagram showing an embodiment in which the present invention also serves as an IDTV system. 7,40 ... Analog-digital converter, 8 ... De-emphasis circuit, 9 ... Synchronization processing circuit, 10 ... Separation circuit, 11
...... Luminance signal inverse gamma correction circuit, 12 ...... Luminance signal noise reduce circuit, 13,14 ...... Luminance signal frame memory, 15 ...... Luminance signal still image processing circuit, 16 ...... Luminance signal moving image processing Circuit, 17 ... Inter-field processing circuit for luminance signal, 18 ... Luminance signal mix circuit, 19 ... Motion detection circuit for luminance signal, 20 ... Time extension circuit, 21 ... Noise reduction circuit for color signal, 22, 23 ... Color signal frame memory, 24 ... Color signal still image processing circuit, 25 ... Color signal moving image processing circuit, 26 ... Color signal inter-field processing circuit, 27
...... Color signal mix circuit, 28 …… Color signal motion detection circuit, 29 …… Color decoding circuit, 30 …… Color signal inverse gamma correction circuit, 31 …… Inverse matrix circuit, 32 …… Digital-analog converter , 34 …… Subtractor, 35 …… Coefficient multiplier, 36 …… Adder, 41 …… Color signal demodulation circuit.

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Toshiyuki Sakamoto 292 Yoshida-cho, Totsuka-ku, Yokohama-shi, Kanagawa Inside the Home Appliances Research Laboratory, Hitachi, Ltd. (72) Takumi Okamura 292 Yoshida-cho, Totsuka-ku, Yokohama-shi, Kanagawa Home Appliances Research Laboratory, Hitachi, Ltd. (72) Ichio Nakagawa, 292 Yoshida-cho, Totsuka-ku, Yokohama, Kanagawa Prefecture Home Office, Home Appliances Research Laboratory, Hitachi, Ltd. (72) Yuichi Ninomiya 1-10-11 Kinuta, Setagaya-ku, Tokyo Japan Broadcasting Engineering Institute Broadcasting Technology Laboratory (72) Inventor Yoshimichi Otsuka 1-10-11 Kinuta, Setagaya-ku, Tokyo Inside Broadcasting Engineering Laboratory Japan Broadcasting Corporation (56) Reference JP-A-62-189893 (JP, A) JP Flat 2-13190 (JP, A) JP-A-2-44986 (JP, A)

Claims (1)

(57) [Claims] 1. A device for decoding a MUSE signal transmitted by band compression using a field and frame offset sampling method that makes a cycle of four fields.
A / D converter for analog-to-digital conversion of SE signal, separating means for separating color signal and luminance signal time-division multiplexed into the MUSE signal, and luminance signal to which the luminance signal is supplied from the separating means Noise reducer, a first frame memory connected to the noise reducer for storing at least one frame of luminance signal, and a first still image processing means for performing still image processing on the luminance signal noise reduced by the luminance signal noise reducer. A first moving image processing means for performing moving image processing, a first motion detecting means for detecting a motion, and a first moving image processing means for detecting a motion from the first still image processing means according to a motion amount from the first motion detecting means. A luminance signal decoding unit including a luminance signal and a first mixing unit that mixes a luminance signal from the first moving image processing unit, and a color signal is supplied from the separating unit. A time axis expanding means for expanding the time axis compressed color signal in the time axis, a color signal noise reducer connected to the time axis expanding means, and one or more color signal frames connected to the color signal noise reducer are stored. A second frame memory, a second still image processing means for performing still image processing on the color signal subjected to noise reducer processing by the color signal noise reducer, a second moving image processing means for performing moving image processing, and motion detection. And a color signal from the second still image processing unit and a color signal from the second moving image processing unit are mixed according to the amount of movement from the second motion detecting unit. A MUSE signal decoding device, comprising: a color signal decoding means including a second mixing means;