JP2003153038A - Video signal processing apparatus - Google Patents

Abstract

PROBLEM TO BE SOLVED: To achieve image correction in which granular noises are not emphasized in an interframe coding (I) image and a forward interframe prediction coding (P) image. SOLUTION: The device consists of an MPEG decoder 2 for decoding an inputted coding video image, a control circuit 14 for outputting a dead band control signal 8 for controlling the dead band of a core ring circuit 4 from coding information 7 outputted from the decoder 2 and a gain control signal 13 for controlling a gain of a gain adjusting circuit 12 for determining an output amplitude of a primary differentiating circuit, and an adder 15 for adding the output video signal of the decoder 2, the output of the circuits 14, 12. In the case of I, P images, the circuit 14 controls the dead band of the circuit 4 extensively and performs control so that the gain of the circuit 12 may increase, thereby improving the feeling of sharpness of a portion of contour boundary without emphasizing the granular noises appearing in the I and P images.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a video for decoding a coded image composed of an intra-frame coded image, a forward inter-frame predictive coded image, and a bidirectional predictive coded image to obtain a video signal. The present invention relates to a signal processing device.

[0002]

2. Description of the Related Art Conventionally, when a moving picture is coded using an intra-frame coded image, a forward inter-frame predictive coded image, and a bidirectional predictive coded image, an intra-frame coded image, a forward frame In the inter-prediction coded image, it is the image that will be used for the next prediction, so the quantization step is finely divided to improve the accuracy. In the bidirectional predictive-coded image, the quantization step is not used next. The average SNR (Signal to Noise R
to improve the audio.

That is, in the intra-frame coded image and the forward direction inter-frame predictive coded image, even the fine noise granularity noise is coded because the quantization step is fine, whereas in the bidirectional predictive coded image, Granular noise of small amplitude is not encoded due to the coarse quantization step.

In addition, an intra-frame coded image, a forward inter-frame predictive coded image, and a bidirectional predictive coded image are used.
In addition, a signal processing method at the time of decoding data band-compressed and encoded by using discrete cosine transform (DCT) with a plurality of adjacent pixels as one block is disclosed in JP-A-11-113.
The one described in No. 000 is known.

This is only for the DC component of the above block,
The decoding device applies a filter in the time axis direction and further processes only bidirectional predictive coded images.

[0006]

In this decoding device, since processing is performed only on the DC component of the block, it is effective for block distortion, but intra-frame coded image,
It has no effect on granular noise due to fine quantization steps in the forward inter-frame predictive coded image.

[0007]

In order to solve the above-mentioned problems, a video signal processing apparatus of the present invention comprises an MPEG decoder for decoding input coded video, and coding method information output from the MPEG decoder. And an image quality correction circuit for performing image quality control on the decoded video signal output from the MPEG decoder based on the output of the control device.

According to the invention, an intra-frame coded image,
The sharpness can be improved without emphasizing the granular noise due to the fine quantization step in the forward inter-frame predictive coded image.

[0009]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The invention according to claim 1 of the present invention is a control for obtaining an image quality control signal from an MPEG decoder for decoding input coded video, and coding method information output from the MPEG decoder. Device and the above MPE
The decoding video signal output from the G decoder is characterized by being provided with an image quality correction circuit for performing image quality control based on the output of the control device, and a filtering process for an input image to the encoding device. Also, it has an effect that the sharpness can be improved without enhancing the noise without requiring the filtering information.

According to a second aspect of the present invention, the image quality correction circuit according to the first aspect is configured to pass a high band of a decoded video signal output from the MPEG decoder, and the high pass filter. A coring circuit that performs coring processing on the output of, and an addition circuit that adds the decoded video signal and the coring circuit,
When the encoding method information indicates an intra-frame encoded image, the encoding method is configured by a control device that performs control to increase the coring amount of the coring circuit, and a filtering process for an input image to the encoding device. ,and,
For the intra-frame coded image with a lot of granular noise without filtering information, the enhancement of the grain noise is suppressed by suppressing the high frequency emphasis, and for the bidirectional predictive coded image with a little grain noise, It has the effect of emphasizing high frequencies and improving sharpness.

According to a third aspect of the present invention, in the control device according to the second aspect, when the coding method information indicates an intra-frame coded image or an inter-frame forward prediction coded image, It is a control device that performs control to increase the coring amount of the coring circuit,
Noise can be removed on the decoding device side regardless of the filtering information on the encoding device side, and high-frequency emphasis can be suppressed even for inter-frame forward predictive encoded images that have the second largest grain noise after the intra-frame encoded image. Has the effect of suppressing the enhancement of the granular noise and enhancing the high frequency band for a bidirectional predictive-coded image with a small amount of granular noise to improve the sharpness.

According to a fourth aspect of the present invention, in the image quality correction circuit according to the second aspect, the high-pass filter is provided.
In addition to the coring circuit, a low-pass filter that passes the low-pass of the decoded video signal, and a first-order differentiation circuit that performs first-order differentiation on the output signal of the low-pass filter,
The control device comprises a gain adjusting circuit for adjusting the output gain of the primary differentiating circuit, an adding circuit for adding the output of the coring circuit, the output of the gain adjusting circuit and the decoded video signal, When the encoding method information indicates an intra-frame encoded image, the coring amount of the coring circuit is increased and the gain of the gain adjusting circuit is increased. In addition, it is possible to compensate for the lack of sharpness of the intra-frame coded image that occurs as a result of suppressing the enhancement of the granular noise of the intra-frame coded image by the coring circuit.

According to a fifth aspect of the present invention, in the control device according to the third aspect, when the encoding method information indicates an intra-frame encoded image or an inter-frame forward prediction encoded image, The configuration is such that the amount of coring of the coring circuit is increased and the gain of the gain adjustment circuit is increased. In addition to the function of claim 3, the intra-coded image in the coring circuit, and , It is possible to compensate for the lack of sharpness of the intra-frame coded image generated as a result of suppressing the enhancement of the granular noise of the inter-frame forward prediction coded image.

FIG. 1 shows an embodiment of the present invention.
3 to FIG. 3 will be described.

(Embodiment 1) FIG. 1 shows a block diagram of a first embodiment of the present invention. In FIG. 1, an input coded video 1 is a video signal coded by an MPEG encoder.

The MPEG decoder 2 receives the input coded video 1
, The MPEG-decoded video signal is output, and the input coded video 1 is MPEG-decoded, the input coded video 1 is an intra-frame coded image, an inter-frame forward prediction coded image, a bidirectional The code information 7 indicating which image of the predictive coded image is output.

The HPF 3 is a high-pass filter which receives the MPEG-decoded video signal output from the MPEG decoder 2 as an input and passes a high-pass component.

When the image type indicated by the code information 7 is an intra-frame coded image, the control device 5 outputs a dead zone control signal 8 for widening the dead zone of the coring circuit 4 described later.

The coring circuit 4 receives the high frequency component of the MPEG-decoded video signal obtained by the HPF 3 and the dead zone control signal 8 and has the input / output amplitude characteristic shown in FIG. When the amplitude of the high frequency component of the video signal is smaller than the amplitude indicated by the dead zone control signal 8, the dead band is provided by setting the output to 0 regardless of the input, and when the amplitude is large, the output increases as the input amplitude increases. It is a circuit having.

The adder 6 outputs the output of the coring circuit 4 and M
This is a circuit for adding the video signal from the PEG decoder 2 to obtain an output video signal 9.

With the above configuration, when the input coded image 1 is an intra-frame coded image with a lot of granular noise,
Since the controller 5 controls the dead zone control signal 8 to widen the dead zone of the coring circuit 4, the enhancement of the small-amplitude high-frequency component of the output video from the MPEG decoder 2 is prevented, while the input encoding is performed. When the video 1 is other than the intra-frame coded image, the control device 5 controls the narrowing of the dead zone of the coring circuit 4, so that the high frequency component is emphasized in the output video of the MPEG2 decoder 2.

As a result, it is possible to suppress the granular noise by suppressing the high frequency enhancement for the intra-frame coded image having a lot of granular noise without the need for the filtering process for the input image to the encoding device and the filtering information. It is possible to suppress the emphasis and enhance the high frequency band for a bidirectional predictive-coded image with a small amount of granular noise to improve the sharpness.

(Embodiment 2) In the video signal processing apparatus according to Embodiment 1 of the present invention, the control for suppressing the enhancement of the small amplitude high frequency component is performed only for the intra-frame coded image. As for the inter-frame forward predictive coded image having a smaller quantization step next to the intra-frame coded image, that is, a large amount of granular noise, the small-amplitude high-frequency component is emphasized. It occurs when the predictive coded image is decoded.

In the second embodiment of the present invention, in order to solve the problem of the first embodiment, a high-amplitude component with a small amplitude is obtained for both the intra-frame coded image and the inter-frame forward prediction coded image. Control the emphasis on.

A second embodiment of the present invention will be described with reference to FIG. The input coded video 1, the MPEG decoder 2, the HPF 3, the coring circuit 4, the adder 6, the code information 7, and the video output 9 are the same as those described in the first embodiment. Is omitted.

When the image type indicated by the code information 7 is an intra-frame coded image or an inter-frame forward predictive coded image, the control device 5 will be described later with reference to a coring circuit 4
The dead zone control signal 8 for widening the dead zone is output.

With the above configuration, when the input coded image 1 is the intra-frame coded image having a lot of granular noise and the inter-frame forward predictive coded image, the control device 5 uses the dead zone control signal 8. Since the dead zone of the coring circuit 4 is controlled to be wide, the enhancement of the small-amplitude high-frequency component of the output video from the MPEG decoder 2 is blocked, while the input coded video 1 is a bidirectional predictive coded image. , The control device 5 controls to narrow the dead zone of the coring circuit 4,
The high frequency component is emphasized in the output image of the PEG2 decoder 2.

As a result, the filtering process for the input image to the encoding device and the small prediction for the intra-frame coded image and the inter-frame forward prediction coded image with a lot of granular noise are not required. The enhancement of the granular noise is suppressed by suppressing the high frequency emphasis of the amplitude, and the high frequency is emphasized for the bidirectional predictive-coded image having a small amount of the granular noise to improve the sharpness. It is possible to further improve the sharpness while suppressing the enhancement of granular noise.

(Embodiment 3) In the video signal processing apparatus shown in Embodiment 1 of the present invention, control for suppressing enhancement of a small-amplitude high-frequency component is performed in the case of an intra-frame coded image. By suppressing the enhancement of the high frequency component even with a small amplitude, the sharpness of the contour boundary portion that is originally desired to be emphasized is easily lost.

Therefore, in the third embodiment of the present invention, the above problem is solved by performing contour enhancement on the contour boundary portion in order to supplement the sharpness of the contour boundary portion.

A third embodiment of the present invention will be described with reference to FIG. Note that, in FIG. 2, components and signals given the same reference numerals as those in FIG. 1 are the same as those in FIG. 1, and detailed description thereof will be omitted.

In FIG. 2, an LPF 10 is a filter for passing a low frequency band to an output video from the MPEG decoder 2, and a primary differentiating circuit 11 is a circuit for performing a primary differential on the output video from the LPF 10. Is.

When the code information 7 which is another output of the MPEG decoder 2 is an intra-frame coded image, the control device 14 outputs the dead zone control signal 8 for widening the dead zone of the coring circuit 4 and, at the same time, will be described later. The gain control signal 13 for increasing the gain of the gain adjusting circuit 12 is output.

The gain adjusting circuit 12 is a gain adjusting circuit for controlling the amplitude of the output of the primary differentiating circuit 11 based on the gain control signal 13.

The adder 15 is a circuit for adding the output video from the MPEG decoder 2, the output of the coring circuit 4 and the output of the gain adjusting circuit 12, and the output video signal 9
To get

Next, the operations of the LPF 10, the first-order differentiating circuit 11, the gain adjusting circuit 12, and the adder 15 will be described with reference to FIG.

In FIG. 4, the waveform A shows the high frequency noise and the contour boundary portion of the output video signal of the MPEG2 decoder 2.

The waveform A is input to the LPF 10 and is subjected to low pass filter processing to obtain a waveform B. In the waveform B, the high frequency noise existing in the waveform A is removed.

The waveform B is input to the primary differentiating circuit 11,
After performing the first-order differentiation processing, the gain adjustment circuit 12 performs gain adjustment to obtain a waveform C. For the waveform C, a change in the inclination of the contour boundary portion of the waveform B is obtained.

The waveform C and the waveform A output from the MPEG decoder 2 are input to the adder 15 and subjected to addition processing to obtain a waveform D. As a result, only the contour boundary portion of the waveform A is emphasized.

The video signal processing device configured as described above is
It works as follows. When the output video of the MPEG decoder 2 is an intra-frame coded image with a lot of granular noise, the control device 14 uses the dead zone control signal 8 to output the coring circuit 4
Control for widening the dead zone of the
Uses the gain control signal 13 to increase the output amplitude of the primary differentiating circuit 11.

On the other hand, when the output video of the MPEG decoder 2 is other than the intra-frame coded image, the control device 14 controls the dead zone control signal 8 to narrow the dead zone of the coring circuit 4 and the gain control signal 13. Using 1
The output amplitude of the secondary differentiating circuit 11 is suppressed more than in the case of the intra-frame coded image.

As a result, the intra-frame coded image output from the MPEG decoder 2 does not require the filtering process for the image input to the coding device and the filtering information, and the enhancement of the granular noise is suppressed, and
The contour boundary portion is emphasized, and when the video signal output from the MPEG decoder 2 is other than the intra-frame coded image, the high frequency region is emphasized to obtain an image with improved sharpness, and thus the contour to be originally emphasized. It is possible to suppress the enhancement of the granular noise without losing the sharpness of the boundary portion.

(Embodiment 4) In the video signal processing apparatus shown in Embodiment 3 of the present invention, the emphasis of the small amplitude high frequency component is suppressed only for the intra-frame coded image, and the contour boundary part is emphasized. However, for the inter-frame forward-prediction coded image that has a finer quantization step next to the intra-frame coded image, that is, there is much grain noise, the small-amplitude high-frequency component is emphasized, resulting in grain noise. Occurs when the inter-frame forward prediction coded image is decoded.

In the fourth embodiment of the present invention, in order to solve the problem of the third embodiment, a high-amplitude component of a small amplitude is generated for both the intra-frame coded image and the inter-frame forward prediction coded image. Is suppressed and the contour boundary part is emphasized.

A fourth embodiment of the present invention will be described with reference to FIG. In FIG. 2, the input coded video 1 and MP
The EG decoder 2, HPF 3, coring circuit 4, adder 15, code information 7, video output 9, LPF 10, first-order differentiating circuit 11, and gain adjusting circuit 12 are the same as those described in the third embodiment. And its description is omitted.

The control device 14 widens the dead zone of the coring circuit 4 when the code information 7 which is another output of the MPEG decoder 2 is an intra-frame coded image or an inter-frame forward predictive coded image. At the same time as outputting the dead zone control signal 8, the gain control signal 13 for increasing the gain of the gain adjusting circuit 12 is output.

When the output video of the MPEG decoder 2 is an intra-frame coded image with a lot of granular noise or an inter-frame forward coded image with a lot of granular noise next, the controller 14 uses the dead zone control signal 8 to perform coring. At the same time as performing control to widen the dead zone of the circuit 4, the control device 1
4 uses the gain control signal 13 to increase the output amplitude of the primary differentiating circuit 11.

On the other hand, when the output video of the MPEG decoder 2 is a bidirectional predictive coded image, the control device 14 controls the dead zone control signal 8 to narrow the dead zone of the coring circuit 4 and the gain control signal 13. Is used to suppress the output amplitude of the primary differentiating circuit 11.

As a result, the intra-frame coded image output from the MPEG decoder 2 or the inter-frame forward coded image output from the MPEG decoder 2 is not required for the filtering process for the image input to the coding device and the filtering information. At times, enhancement of granular noise is suppressed and contour boundary portions are enhanced, and when the video signal output from the MPEG decoder 2 is a bidirectional predictive-coded image, a video close to the video signal from the MPEG decoder 2 is obtained. As a result, it is possible to improve the sharpness of the contour boundary portion that is originally desired to be emphasized while suppressing the enhancement of the granular noise more than in the third embodiment.

[0051]

As described above, according to the present invention, an intra-frame coded image with a fine quantization step or a frame without a filtering process and filtering information for an input image to a coding device is required. The advantageous effect that the sharpness of the contour boundary portion can be improved while suppressing the enhancement of the granular noise of the inter-forward coded image is obtained.

[Brief description of drawings]

FIG. 1 is a block diagram showing a video signal processing device according to a first embodiment of the present invention.

FIG. 2 is a block diagram showing a video signal processing device according to a second embodiment of the present invention.

FIG. 3 is a diagram showing input / output characteristics of the coring circuit of the present invention.

FIG. 4 is a waveform diagram showing the operation of the contour boundary emphasis circuit of the present invention.

[Explanation of symbols]

1 Input coded video 2 MPEG decoder 3 HPF 4 Coring circuit 5 control device 6 adder 7 Code information 8 Dead zone control signal 9 Output video signal 10 LPF 11 First-order differentiation circuit 12 Gain adjustment circuit 13 Gain control signal 14 Control device 15 adder

Claims (5)

[Claims] 1. An M for decoding an input encoded video
A PEG decoder, and a control device for obtaining an image quality control signal from the encoding method information output from the MPEG decoder,
A video signal processing device comprising an image quality correction circuit for performing image quality control on the decoded video signal output from the MPEG decoder based on the output of the control device. 2. A high-pass filter that allows the image quality correction circuit to pass a high band of a decoded video signal output from the MPEG decoder, and coring that performs coring processing on the output of the high-pass filter. A circuit and an adder circuit for adding the decoded video signal and the output of the coring circuit, and when the encoding method information indicates an intra-frame encoded image, the coring amount of the coring circuit 2. The video signal processing device according to claim 1, wherein the video signal processing device comprises a control device that performs control to increase 3. The control device controls to increase the coring amount of the coring circuit when the encoding method information indicates an intra-frame encoded image or an inter-frame forward prediction encoded image. The video signal processing device according to claim 2, wherein the video signal processing device is a control device. 4. The image quality correction circuit includes, in addition to the high pass filter, the coring circuit, a low pass filter that passes a low band of the decoded video signal, and an output signal of the low pass filter. A first-order differentiating circuit for performing a first-order differentiating operation, a gain adjusting circuit for adjusting an output gain of the first-order differentiating circuit, an output of the coring circuit, an output of the gain adjusting circuit, and the decoded video signal. When the encoding method information indicates an intra-frame encoded image, the control device increases the coring amount of the coring circuit and sets the gain of the gain adjusting circuit. The video signal processing device according to claim 2, wherein the video signal processing device is configured to be raised. 5. The control device increases the coring amount of the coring circuit when the encoding method information indicates an intra-frame encoded image or an inter-frame forward prediction encoded image, and The video signal processing device according to claim 4, wherein the gain adjusting circuit is configured to increase the gain.