JPS63204983A - Noise suppressing method for video signal - Google Patents

Abstract

PURPOSE:To sufficiently suppress the noise of a still picture by recording a still picture code to show whether to be the still picture or not, while being multiplexed to the video signal, dn a recording medium, and mixing the video signal, obtained from the recording medium together with the signal, obtained by delaying the said video signal by M-time (M is a natural number) of one frame, at a mixing ratio corresponding to the still picture code. CONSTITUTION:A still picture code generation circuit 10 comes to a state, e.g., '1', to show to be the still picture, when the output data of a motion detection circuit 7 is less than a prescribed value, and a multiplex circuit 4 superimposes the still picture code onto a prescribed position within the vertical blanking period of the video signal, by a horizontal and a vertical synchronizing signals (h) and (v), separated from the video signal by a synchronizing separation circuit 3. Thus, the still picture code to show whether it is the still picture or not, being multiplexed to the video signal, is recorded on the recording medium, and the video signal, obtained from the recording medium is mixed together with the signal, obtained by delaying the said video signal by M-time (Mis the natural number) of one frame, at the mixing ratio corresponding to the still picture code, obtained from the recording medium, and the noise is suppressed.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a method for suppressing noise in a video signal, and more particularly to a method for suppressing noise in a video signal recorded on and reproduced from a recording medium such as a recording disk or a magnetic tape.

1 and I A video signal is a signal that indicates information on each part of a video at a frame period, and the autocorrelation between frames is very strong. on the other hand,
Noise components contained in video signals generally have almost no autocorrelation, so if the video signal is temporally averaged for each frame period, the energy of the signal components will hardly change and the energy of the noise components will decrease. only decreases. However, real images have movement, and if the average is simply applied, the moving image portion will become temporally blurred.

Based on this knowledge, when a video signal is recorded on recording medium B and played back, the video signal and this video signal are delayed by a period of time M times one frame (M is a natural number) on the playback side. A noise suppression method has already been devised in which noise is suppressed by mixing signals with a signal at a mixing ratio corresponding to the instantaneous level difference (movement) between these two signals. In this conventional noise suppression method, noise is reduced by increasing the ratio of the delayed video signal and decreasing the ratio of the undelayed video signal (hereinafter referred to as the current video signal) in the still part. In this section, blur is reduced by reducing the ratio of the delayed video signal and increasing the ratio of the current video signal.

In such conventional noise suppression methods, if there is noise even in a still image, a level difference will occur between the video signal and the delayed video signal, so if motion detection is performed based on the instantaneous level difference, the entire area of the still image may not be completely covered. This has the disadvantage that it is not detected as a static portion, the ratio of the delayed video signal is not large enough, and the noise level cannot be made sufficiently small.

Therefore, it may be possible to detect on the playback side whether or not the image based on the video signal obtained from the recording medium is a still image, and in the case of a still image, to make the ratio of the delayed video signal sufficiently large. In this case, the scale of the circuit on the playback side becomes large, which is not desirable.

An object of the present invention is to provide a method for suppressing noise in a video signal, which can suppress noise in a still image by a considerable amount without increasing the circuit scale on the reproduction side.

A method for suppressing noise in a video signal according to the present invention involves multiplexing a still image code indicating whether or not it is a still image onto a video signal and recording it on a recording medium, and adjusting the mixing ratio according to the still image code obtained from the recording medium. It is characterized in that a video signal obtained from a recording medium and a signal obtained by delaying this video signal by M times one frame (M is a natural number) are mixed to suppress noise.

Friend-blin-1 Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

In Fig. 1, a video signal output from a VTR (video tape recorder) etc. is input to an A/D converter via an input terminal 1.
It is supplied to an (analog-to-digital) converter 2, a synchronous separation circuit 3, and a multiplex circuit 4. The A/D converter 2 is supplied with a clock pulse a of a predetermined frequency output from the pulse generating circuit 5. A/D converter 2
, the video signal is sampled by the clock pulse a, and the obtained sample values are sequentially converted into digital data. The output data of this A/D converter 2 is supplied to a one frame delay circuit 6 and a motion detection circuit 7. The one-frame delay circuit 6 is connected to a frame memory 8 having a storage capacity capable of storing one frame's worth of data, and after one frame's worth of data is sequentially written to the frame memory 8 by a clock pulse a. It is comprised of a memory controller 9 that performs address control of the frame memory 8 so that written data is read out sequentially in the order in which it was written. This one frame delay circuit 6 delays the output data of the A/D converter 2 by one frame period.

Further, the motion detection circuit 7 is configured to output data corresponding to the absolute value of the difference between the output data of the A/D converter 2 and the output data of the one-frame delay circuit 6 as data indicating motion. The output data of this motion detection circuit 7 is supplied to a still picture code generation circuit 10. The still picture code generation circuit 10 is supplied with a vertical synchronization signal separated from the video signal by the synchronization separation circuit 3. The still image code generation circuit 10 determines that the image is a still image when the output data of the motion detection circuit 7 is equal to or less than a predetermined value during one frame period, that is, the period from when the vertical synchronizing signal once disappears until it is generated again. A state indicating that, for example, 1
'', and when the output data of the motion detection circuit 7 exceeds a predetermined value within one field period, the 1-bit still image code indicating that it is not a still image, that is, for example, "0", is transferred to the next one field period. This still image code is supplied to a multiplex circuit 4. The multiplex circuit 4 outputs the horizontal and vertical synchronization signals separated from the video signal by the synchronization separation circuit 3 and V, the still picture code is superimposed at a predetermined position within the vertical blanking period of the video signal.

The video signal multiplexed with the still image code by the multi-m circuit 4 is supplied to the frequency modulator 11, and is, for example, 8.1 MH.
Modulation processing of the FM carrier wave of z is performed to generate an FM signal. This FM signal is supplied to the optical modulator 12, and the transmittance of this optical modulator 12 is controlled. As a result, the Radhe light generated by the laser light source 13 is intensity-modulated by the optical modulator 12, then expanded by the collimator lens 14, and then by the converging lens 15 onto the recording surface of the master disk 16 with a diameter of approximately 1 μm.
The photoresistor and the like constituting the recording surface are exposed by converging the light spot.

On the other hand, the master disk 16 is rotated by a motor 17 and a servo loop (not shown) at a speed that depends on the relative position of the light spot in the radial direction of the master disk 16, and at the same time, at the same time, the number of rotations per rotation in the radial direction is It is transferred at a rate of about 2 μm.

Further, the converging lens 15 is controlled by a servo loop so as to always converge the laser beam onto the recording surface of the master disk 16. Since these are not particularly necessary for the explanation, they are omitted in this figure.

A reproducing apparatus for reproducing information from a master disc on which information has been recorded in this manner or from a duplicate disc made by transferring the information onto plastic using a mold produced from the master disc will be explained with reference to FIG.

Disk 2 rotationally driven by spindle motor 20
The recorded information of No. 1 is read by the optical pickup 22. The pickup 22 includes a laser diode, an objective lens, a focus actuator, a tracking actuator, a photodetector, and the like.

Laser light emitted from a laser diode in the pickup 22 is focused by an objective lens onto the recording surface of the disk 21 to form a light spot. The reflected light from this light spot is incident on a photodetector in the pickup 22, and information is read. Further, the position of the optical spot in the disk radial direction is controlled by a tracking actuator and a servo loop (not shown) in the pickup 22 so that the optical spot formed on the recording surface of the disk 21 is located on the recording track. Ru. Furthermore, the objective lens in the pickup 22 is controlled by a focus actuator and a servo loop to focus the laser beam onto the recording surface of the disk 21.

Since these are not particularly necessary for the explanation, they are omitted in this figure.

An RF (high frequency) signal outputted from the pickup 22 is supplied to an FM demodulation circuit 23 to demodulate the video signal. This video signal is supplied to an A/D converter 24, a synchronous separation circuit 25, and a still image code reading circuit 26. The still picture code reading circuit 26 generates, for example, a horizontal and vertical synchronization signal separated from the video signal by the synchronization separation circuit 25 and a timing signal generated at a timing corresponding to the position where the still picture code is superimposed by (1). , is configured to read and hold the still image code using this timing signal. The still image code obtained by the still image code reading circuit 26 is supplied to a noise reduction circuit 27.

On the other hand, in the Δ/D converter 24, the video signal is sampled using the clock pulse b of a predetermined frequency output from the pulse generating circuit 28, and the obtained sample values are sequentially converted into digital data. This A
The output data of the /D converter 24 is supplied to a noise reduction circuit 27. Noise reduction circuit 27
, the output data of the A/D converter 24 is supplied to a subtraction circuit 29 and a subtraction circuit 30. Subtraction circuit 29
The output data of is supplied to the one frame delay circuit 31 and simultaneously supplied to the output terminal 32.

The 1-frame delay circuit 31 is composed of a frame memory 32 and a memory controller 33, similar to the 1-frame delay circuit 6 in the device shown in FIG. The output data of the subtraction circuit 29 is delayed by one frame period by this one-frame delay circuit 31, and then supplied to the subtraction circuit 30.
Difference data ε obtained by subtracting the output data of the frame delay circuit 31 is calculated. The output data of the subtraction circuit 30 is supplied to ROMs (read only memories) 34 and 35 as address data. At each address in the ROM 34, difference data ε corresponding to the address value and data f+ (ε) having a relationship as shown in FIG. 3 are stored in advance. Further, each address of the ROM 35 has a relationship as shown in FIG. 4 with difference data ε corresponding to the address value, and data f+
Data f2 (ε) having a larger absolute value than (ε) is stored in advance.

The data read from these ROMs 34 and 35 are supplied to two input terminals of a changeover switch 36, respectively. A still image code is supplied to the changeover switch 36 as a control input. The changeover switch 36 selectively outputs the data read from the ROM 34 when the still image code is "0", and outputs the data read from the ROM 34 when the still image code is "1".
It is configured to selectively output the data read from 35. The output data of this changeover switch 36 is
The signal is supplied to a subtraction circuit 29 and subtracted from the output data of the A/D converter 24.

The noise reduction circuit 27t in the above configuration,
It has a recursive filter configuration, and the difference data ε output from the equalization circuit 30 is the output data of the A/D converter 24, that is, the data corresponding to the instantaneous level of the current video signal, and the output data of the noise reduction circuit 27. This is the difference between the data obtained by delaying the data by one frame period. Data f+ (ε) and fz corresponding to this difference data ε
(ε) is output from each of the ROMs 34 and 35.

These data f+ (ε) and fx (ε) can be said to be data obtained by multiplying the difference data ε by a coefficient corresponding to the difference data ε. These data f+
(ε) and fz (ε) is output from the changeover switch 36 according to the still image code and subtracted from the output data of the A/D converter 24 in the subtraction circuit 29, so that the current video signal and the delayed video signal are However, the data mixed with the still image code at a mixing ratio suitable for the video signal content is output from the subtraction circuit 29 and becomes the output of the noise reduction circuit 27.

Here, the absolute value of data fz (ε) is data f+
(ε), and when the still image code is “1”, that is, when it is a still image, this data f
2 (ε) is subtracted from the output data of the A/D converter 24, the ratio of the delayed video signal in the output data of the frame reduction circuit 29 becomes large, and noise is sufficiently suppressed. Furthermore, when the still image code is 0'', that is, when it is not a still image, data f+ (ε) is subtracted from the output data of the A/D converter 24, increasing the ratio of the current video signal and preventing the occurrence of blur. be done.

In the above embodiment, the still image code is a 1-bit code, but the still image code may be a multiple-bit code, which enables highly accurate switching control.

The above is a guide to recording and playing back NTSC video signals.
A 1. :Twit theory 11+1 L/tc is M
US E (Multiple Sub-Nyquis
The present invention can also be applied to the case of recording and reproducing high-definition television signals whose bandwidth has been compressed using the tSampling Encoding method. In addition, in this case, MU
By controlling the J detection section in the SE decoder using the still image code, it is possible to prevent interference (occurrence of glare) caused by erroneously detecting a still portion as a moving image portion.

As detailed above, the method for suppressing noise in a video signal according to the present invention involves multiplexing a poly still image code on a video signal to determine whether it is a still image and recording the same on a recording medium. Since the noise is suppressed by mixing the video signal obtained from the recording medium with the signal obtained by delaying this video signal by M times one frame (M is a natural number) at a mixing ratio according to the still image code, A separate detection means for detecting whether or not it is a still image is provided on the playback side.
Still images can be detected even without the provision, and noise in still images can be sufficiently suppressed without increasing the circuit scale on the reproduction side.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing a recording device that multiplex-records still image codes using the method according to the present invention, and FIG. 3 is a graph showing the data stored in the ROM 34 in the device shown in FIG. 2, and FIG. 4 is a graph showing the data stored in r(0M35) in the device shown in FIG. 2. Applicant Pioneer Corporation Agent Patent Attorney Motohiko Fujimura Kuchitoge

Claims (1)

[Claims] A method for suppressing noise in a video signal recorded on a recording medium and reproduced, wherein the video signal and the video signal are divided into one frame of N(
detecting the amount of motion of the video obtained by the video signal based on the instantaneous level difference between the signal and the signal obtained after delaying the video signal by a period of time (N is a natural number); A still image code of at least 1 bit indicating whether or not it is a still image is generated depending on whether or not the image exists, the still image code and the video signal are multiplexed and recorded on a recording medium, and the The still image code is obtained together with the video signal from a recording medium, and the video signal and a signal obtained by delaying the video signal by a period of time M times one frame (M is a natural number) are transmitted according to the still image code. A video signal noise suppression method characterized by suppressing noise by mixing at a mixing ratio.