JPS61248691A - Signal processing circuit for high definition television receiver - Google Patents

Abstract

PURPOSE:To obtain a normal freeze picture when an animation is freezed by cycling a signal of one frame on which a motion adapting process is applied in a freeze mode within a memory and using the signal of the one frame on which the motion adapting process is applied within the memory. CONSTITUTION:In a normal mode, the signal from a switch circuit 12 is led to field memories 14 and 15 having share of four fields by a switch circuit 31. On the other hand, in the freeze mode, the switch circuit 31 is connected as shown at a figure and also, the switch circuits 12 and 13 are terminated switching operations which are controlled by a sub-sampling clock signal from an input terminal 7 after one frame and are connected as shown at the figure. To the memories 14 and 15, the output signal of a mixer 21 on which the motion adapting process is applied is always led and also, to a sub-sample filter 16 and to two-dimentional filters 17 and 18, the signal in which the output of the mixer 21 from the second field memory 15 is delayed by one frame is led. Thereby, the signal of one frame before on which the motion adapting process is applied is always feedbacked to the mixer 21 and the freeze picture can be obtained.

Description

[Detailed description of the invention] [Field of application of the invention] The present invention provides a band-compressed high-definition television signal.

to the original broadband television signal.

Related to TV receivers, particularly suitable for this TV receiver.

The present invention relates to a signal processing circuit suitable for providing a freeze function.

[Background of the invention]

Practical application capable of transmitting broadband high-definition television signals.

This is an example of a method for compressing and transmitting a level band.

NHK Giken Monthly, Volume 27, No. 7, 1984.
.

MUSE (MUSE; Multtple 5ub
-Nyquist 3amplin! IEnoodin
! I) There is a method.

As described in this document, this system transmits four fields of broadband high-definition television signals.

A method that performs sub-Nyquist sampling, which in principle compresses the band to about 1/4.

It is a formula.

FIG. 2 shows an embodiment of a decoder section of a receiver that returns a high-definition television signal band-compressed by the Muse method (hereinafter referred to as a Mease signal) to the original wide-band high-definition television signal.

The feature of this embodiment is that it uses field memories 14, 15, 23 that delay a signal obtained by alternately interpolating the previous signal and sampling data by one field;
By interpolating images for four fields, the original broadband television signal is restored. This decoder will be explained below.

In Figure 2, 1 is the muse signal input terminal, 2 and 3
are output terminals for horizontal synchronization (HD) and vertical synchronization (VD), respectively, and output terminals 4 to 6 are output terminals for broadband R, G, and H primary color signals, respectively.

The mean signal from the input terminal 1 is converted into a digital signal by an A/D converter 9 and guided to a noise reduction circuit (hereinafter referred to as NR circuit) 11 and a synchronization processing circuit 10. The synchronization processing circuit 10 detects each control signal from the muse signal, and also detects a sub-sample clock signal and an HD signal for resampling the muse signal.

Outputs VD signal. - The muse signal led to the NR circuit 11 is subjected to noise reduction using the correlation with the signal from the second field memory 15 two frames before. The noise-reduced signal is switched in the first switch circuit 12 using the sub-sample clock from the input terminal 7. Here, the signals of one frame before and two frames before, which are interpolated between frames, are led to the output of the second field memory 15, and the second switch circuit 13 sends the signals of the previous frame and the second frame, similarly to the first switch circuit 12, to the second field memory 15. The other input terminal of the first switch circuit 12 is switched by the sample clock.
The signal from the previous frame is guided to the NR circuit 11, and the signal from two frames before is introduced to the NR circuit 11. Therefore, 1st and 6th.

The output of the switch circuit 12 is from the NR circuit 11.

current muse signal and a second field memory.

The signal from 1b one frame ago or the interpolated signal.

The number is led. This lump Miz signal and one frame.

The interpolated signal is the first field memo and the subsample filter 16th.

1 two-dimensional filter 17. First Fi.

- The signal led to the field memory 14 is one field.

This is delayed and guided to the second field memory 15.

This further delays the input terminal 8 by one field.
Motion compensation is performed using motion vector signals from.

A correction will be made. Therefore, the second field memory 1
An interpolated signal of one frame before which has been subjected to motion compensation is introduced to the output of No. 5. On the other hand, the signal led to the sub-sample filter 16 is filtered by a sub-1: sample cross-circuit to extract only the current muse signal. In addition, the first two-dimensional filter 1
The signal obtained by interpolating the current muse signal derived from 7 and the previous frame signal is subjected to vertical and horizontal two-dimensional filter processing. Similarly, the signal from the second field memory 15 in which the previous frame and the previous frame are interpolated is processed by the second two-dimensional filter 18. 19 is a low-pass filter (hereinafter referred to as LPF), and sub-sample filter 1
The current muse signal from 6 is subjected to horizontal filter processing and is led to a mixer 21 as a moving image signal. -10,000, 20
is a quasi-static processing circuit, the first two-dimensional filter 17
Still image processing is performed using the signals from the second two-dimensional filter 18 and the signals from the second two-dimensional filter 18 are introduced to the mixer 21 as still image signals.

In the mixer 21, a motion detection circuit 22 detects the amount of motion between one frame or two frames from the low-frequency components of each filter 16, 17, and 18, and a motion amount is detected from a temporal filter 26 that temporally expands this amount of motion. Based on the signal, motion adaptive processing is performed to determine the mixing ratio of the moving image processed signal and the still image processed signal. 25
゜ is an interpolation filter, and the movement from the mixer 21 is applied.

Field interpolation is performed between the ratio-processed signal and the signal from the third field memory 24 from 1 field before, and the signal is returned to the original broadband signal. .

26 is a chroma decoding circuit, and the time pressure is 1/4.

and in the line sequence of color difference 1n CW and CN.

Chroma signal time-division multiplexed on luminance signal Y.

Return to the original timeline. 27 and 28 are D/A converters, in which digital signal processing is performed, respectively.

The luminance signal Y and chroma signals CW and CN are analog.

It is converted into a log signal 15, and the original signal is converted into a log signal in a matrix circuit 29.

The signals are returned to broadband primary color signals R, G, and H. this.

In this way, in the embodiment shown in FIG. 2, the Muse signal, which has been band-compressed by 11 degrees, is restored to the original broadband television signal.

As mentioned above, in order to perform the process of interpolating images for four fields, this decoder has four fields consisting of the first and second field memories 14 and 15 as in the embodiment shown in FIG.
Requires field memory for one field. Therefore, by using the memory for these four fields, in the freeze mode, the first and second switch circuits 12 and 13 are forcibly connected to the black circles, contrary to the illustration, and the second field memory 15 A frozen image can be obtained by returning the output of the first field memory 14 to the long sword. However, as mentioned above, in the decoder, the data is transmitted between one frame or two frames.

A motion adaptive process is performed in which the amount of motion is detected and the mixing ratio of the still image processed signal and the video processed signal is changed according to this amount of motion. is not applied. This is because the muse signal is a signal obtained by offset sampling between frames, so the motion detection in the motion detection circuit 22 is mainly based on the difference between two frames. This is because it cannot be detected. Therefore, when a moving image is frozen, moving parts between frames are interpolated as they are, causing problems such as deterioration of the resolution of the frozen image.

[Purpose of the invention]

An object of the present invention is to convert band-compressed high-definition television signals, such as Muse signals, to the original wideband signal.

7. It is an object of the present invention to provide a signal processing circuit that can obtain a normal freeze function in a decoder that returns to a Levi signal.

[Summary of the invention]

In order to achieve the above object, in the present invention, the current muse signal or the signal before the seam is interpolated between frames in the previous stage of the field memory for 4 fields used in the decoder. A switch circuit is provided to switch between the current muse signal or the signal obtained by interpolating the current muse signal and the signal one frame before. ,−
7. In the freeze mode, the motion-adapted signal is guided to the field memory described above to obtain a freeze signal.

[Embodiments of the invention]

An embodiment of the present invention will be described below with reference to FIG.

In FIG. 1, 50 is an input terminal for a freeze mode signal, 61 is a switch circuit that can be switched by the freeze mode signal, and the others are the second embodiment.

The same 0 normal state λ switch circuit 61 is shown as I.

are connected in reverse, field memo 1J14゜15
A signal from the first switch circuit 12 is conducted to the first switch circuit 12 .

On the other hand, in the freeze mode, the switch circuit 61 is connected in the same way as shown in the figure, and the first and second switch circuits 12 and 13 stop switching operations based on the sub-sample clock signal from the input terminal 7 after one frame. , illustrated.

are connected in the same way. As a result, the field memory 14.1!5 contains the mixer 21 which has been subjected to motion adaptive processing.
The output signal of sub-sample filter 1 is always guided, and
A signal obtained by delaying the output signal of the mixer 21 of the second field memory 15 by one frame is introduced to the mixer 21 and the two-dimensional filters 17 and 18. As a result, the motion adaptively processed signal of the previous frame is always fed back to the mixer 21, and the frozen image? Obtainable. In particular, in this embodiment, since the long sword signals to each filter 16, 17, 18 guided to the motion detection circuit 22 are the same, the amount of motion is basically not detected.

In the mixer 21, a still image is processed in a quasi-still circuit 21.

Only the signal is output and once subjected to motion adaptive processing.

There is no need to process the motion adaptation again.

To obtain a normal image that does not cause image quality deterioration due to flicker between frames even if the image is frozen.

I can do it. However, in this case, freeze mode.

After that, the gain of the quasi-static circuit 20 is set to 1° with a delay of one frame, or a switch circuit is provided, and the quasi-static circuit 20 is set to 1°.

Is it necessary to route the input of the static circuit 20 to the mixer 21? 1
In the above description of the embodiment of the present invention shown in FIG. 1, the motion correction using the motion vector signal from the input terminal 8 is not stopped in the freeze mode. However, since the signals led from the mixer 21 to the field memories 14 and 15 have already undergone motion compensation, the motion vector signal must be cut off or the motion compensation in the second field memory 15 must be stopped. is desirable.

Furthermore, in the above explanation, the motion detection in the motion detection circuit 22 is not stopped in the freeze mode, but as described above, the motion detection is conducted to the mixer 21 in the freeze mode.

The signal received has already been subjected to motion adaptive processing.

Therefore, motion detection is not required. Therefore, f.

Stop or mix motion detection when in Lease mode.

Forcibly setting the mixing ratio in the container 21 to only still images.

This causes an image to be generated due to a malfunction in the motion detection circuit 22.

It is possible to suppress frugality. However, this.

Stopping the motion detection circuit 21 or mixing in the mixer 21.

It is free to force the ratio to be still images only.

It is desirable to do this one frame after the switch 31 is connected as shown in the figure by the mode signal.

stomach.

FIG. 3 shows another embodiment of the present invention. .

One embodiment of the present invention shown in FIG. 3 is similar to one embodiment of the invention shown in FIG.

The difference from the example is that a delay circuit 62 is provided. The amount of delay of this delay circuit 32 corresponds to the amount of delay from the input end of the sub-sample filter 16 or two-dimensional filter 17, 18 to the output end of the mixer 21. It is possible to correct the amount of delay for each filter that occurs when changing the time, making it easy to change the time.

There will be no time lag due to However, this amount of delay.

the first or second fee by the amount of delay due to path 32.

The delay amount of the field memories 14 and 15 is reduced, and the first.

From one input terminal of the switch circuit a first and a first second field memory 14.15 and a second switch.

The delay to the other input terminal as it is routed through the circuit.

Do I need to set it so that the extension amount is 1 frame?

Ru.

The embodiment shown in FIG. 3 is the embodiment shown in FIG.

As mentioned in the explanation, the delay circuit 33.54 which delays the 7 reset signal guided to the first and second switch circuits 12.13 and the second field memory 15 for motion compensation by one frame is provided. An example is shown below. Similarly,
A circuit 36 is provided that delays the freeze signal from the switch circuit 35 and the input terminal 38 by one frame, and when in the freeze mode, the signal indicating the forced standstill mode from the input terminal 37 is guided to the mixer 21 with a one frame delay. 12 of the still image processed signal from the quasi-still processing circuit 20
- An example of outputting the following is shown. Furthermore, the gain of the quasi-static circuit 20 is set to 1 in the 1-freeze mode.

Instead, a switch circuit 39 is provided to guide the input signal of the quasi-static RB circuit 20 to the mixer 21. in this case,
As a signal led to the mixer 21 through the switch circuit 39, in addition to the input signal of the quasi-static circuit, one first or second two-dimensional filter i7, 1B (F).

It may be an input signal.

FIG. 4 shows an embodiment of the present invention that is different from FIGS. 1 and 2.

Q: A feature of the embodiment shown in FIG. 4 is that a switch circuit 39 that is switched by a 7-res signal from an input terminal 41 is newly provided after the mixer 21. This switch circuit 39 outputs the signal from the mixer 21 in normal times, and outputs the signal from the second field memory 15 in the freeze mode. Therefore, in the 17 reset mode, signals subjected to one motion adaptive processing are sent to the two switch circuits 31 and 39 and the first and second field memories 14.
．． It will go around a loop consisting of 15, and freeze using the signal in this loop.

By composing images, there is no deterioration in image quality.

A normal image can be obtained. Also, this fruit.

On the downstream side, the signals used for frozen images are mixed.

Since it does not pass through the combined volume 21, the movement guided to the mixer 21 can be arbitrary, and the movement detection circuit 22 is completely frozen.

Especially when in mode, stop or force one movement amount.

There is no need to set it to still image mode.

FIG. 5 shows another embodiment of the present invention.

Embodiment G in FIG. 5, and Embodiment G in FIG. 4.

As in the embodiment shown in FIG. 3, a delay circuit 32 is provided to correct the delay time of each filter that occurs when switching between normal operation and freeze mode, thereby suppressing time lag due to switching. However, in this embodiment, unlike the embodiment in FIG. It is desirable to match the timing of the signal introduced from the mixer 21 with the signal introduced from the mixer 21.

Figure 6 shows a third example with a different circuit configuration from the embodiments in Figures 1 to 5.
An embodiment in which the present invention is applied to a decoder section in which the field memory 24 is reduced will be shown.

This embodiment is characterized by a switch circuit 40 that switches between the signal from the second field memory 1] 15 and the signal from the mixer 21 subjected to motion adaptive processing, as well as the signal from the first field memory 14. A switch circuit 48 is provided to switch between this signal and the signal from the adder 47 which has undergone still image processing, and is switched between the freeze mode and the normal mode. The operation of this embodiment will be explained below.

In FIG. 6, reference numeral 44 indicates a sixth two-dimensional filter having the same characteristics as the first and second two-dimensional filters 17 and 18. The IF4 signal interpolated with the signal is derived and subjected to two-dimensional filter processing. This two-dimensional filtered signal is subjected to static IL image processing in a static circuit 45, and is added to the 1H previous signal 15 from the 1H delay circuit 46 in an adder 47. Normally there are two switch circuits 4]].

48 is connected to the side opposite to that shown in the figure, and receives the signal that has been motion adaptively processed using the current signal from the mixer 21 and the signal from one frame before, and the signal from the adder 47 for one field before and three fields before. The still image processed signal is guided to the interpolation filter 25.

The interpolation filter 25 performs field interpolation processing using only the motion adaptively processed signal from the switch circuit 40 for moving images, and also using the still image processed signal from the switch circuit 48 for still images. Figures 1 to 5
It is possible to obtain the same broadband television signal as in the embodiment shown in the figure. In freeze mode, two switch circuits 40 and 48
are connected as shown in the figure, and the output signal of the second field memory 15 and the output signal of the first field memory 14 are guided to the interpolation filter 25 and subjected to field interpolation processing. As a result, a normal frozen image similar to that of the embodiment of the present invention shown in FIG. 4 can be obtained.

FIG. 7 shows another embodiment of the present invention.

, 16. The embodiment of FIG. 7 is the embodiment of FIG. 6 with two delay circuits 32, 49 and a first field memo.

A circuit 49 for delaying the output signal of step 14 is installed.

As in the embodiment shown in FIG. 5, during normal operation.

Each fee that occurs when switching to Lease mode.

When the delay time due to the router is corrected and cut-off is made.

This makes it possible to suppress gaps. In this case, slow.

The delay time of the detour path 49 is almost the same as that of the delay circuit 32.49.

The same value is selected.

[Effect 1 of the Invention 10 According to the present invention, for example, it is possible to circulate one frame worth of motion adaptive processing in the memory during freeze mode, and the one frame worth of motion adaptive processing in this memory can be circulated in the memory. By applying a signal of 1J, it is possible to obtain a normal frozen image in which interpolation errors or resolution deterioration caused by still image processing are suppressed when a moving image is frozen.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing an embodiment of a signal processing circuit according to the present invention, and FIG. 2 is a Miez type signal processing circuit. FIG. 3 is a circuit block diagram showing a decoder. In the embodiment of the present invention shown in FIG. 1, time lag correction is performed by switching between normal mode and 7° lease mode. Figure 4 is a block diagram of the circuit that made this possible. FIG. 5 is a circuit block diagram showing another embodiment. In the embodiment of the invention of FIG. 4, the normal mode. Compensation for time lag by switching between and freeze mode. Figure 6 is a block diagram of the circuit that made this possible. FIG. 7 is a circuit block diagram illustrating an embodiment using a light source. - Time required for switching between mode and 7-lead mode. FIG. 2 is a block diagram of a circuit that enables error correction. . 9...A/D converter 10...Synchronization processing circuit 11...Noise reduction circuit 12.13, 31.55, 39.40.48...Switch circuit 14...First field memory 15... Second field memory 16... Sub sample filter 17.18.44... Two-dimensional filter 19... LPF for moving image 20... Quasi-static circuit 21... Mixer 22... -Motion detection circuit 23...Temporal filter 24...Third field memory 25...Interpolation filter 26...Chroma decoder 27, 28...D/A converter 29...Matrix circuit 30 , 58 , 42 . . . Freeze signal input terminal 3
2, 43, 49...Delay circuit 55 + 34 + 56 r 41...1 frame delay circuit for freeze signal 45...Still processing circuit 46...1H delay circuit

Claims (1)

[Claims] A signal processing circuit for a high-definition television receiver that converts a high-definition television signal band-compressed by a sub-Nyquist sampling method that goes around in 1 and 4 fields into the original wide-band television signal, which converts at least a high-definition band-compressed signal into the original wideband television signal. A delay circuit that delays a television signal by four fields, a video processing circuit that processes a band-compressed high-definition television signal as a moving image, a still-image processing circuit that processes a band-compressed high-definition television signal as a still image, and a band compression circuit. A motion detection circuit detects the amount of motion between one frame or two frames of a high-definition television signal, and the amount of motion detected by the motion detection circuit is used to detect the signal from the video processing circuit and the still image processing circuit. It is equipped with a mixer that controls the mixing ratio of signals, and a switch circuit provided before and after the delay circuit, and in the freeze mode, the switch circuit causes the mixer connected to one input terminal of the switch circuit to A signal processing circuit for a high-definition television receiver, characterized in that the output signal is guided to the input of a delay circuit.