JP2004140863A - Motion picture decoding apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To solve a problem that a processing quantity is increased or a processing time is wasted by implementing post-filtering to a decoded image conventionally without fail to perform filtering also to a portion not to be post-filtered such as a still picture portion with no change from a preceding frame, and also to solve disadvantage that not only the processing quantity is increased but also a picture quality is deteriorated by filtering without referring to an encoding attribute of blocks so as not to perform filtering suited to a generated noise. <P>SOLUTION: The motion picture decoding apparatus is provided with a filter means for decoding an image encoded for each block and changing pixel values with block boundaries in between as pixel values after filtering in order to smoothly connect the block boundaries in the decoded image, and a frame memory means for storing the decoded image after filtering. The decoded image after filtering is stored in the frame memory means as the preceding frame and when obtaining a decoded image of a present frame, a block is read out of the decoded image after filtering. <P>COPYRIGHT: (C)2004,JPO

Description

The present invention decodes coded data that has been subjected to orthogonal transform / quantization / encoding by dividing into blocks, and reduces block distortion and mosquito noise occurring in a decoded image obtained by inverse quantization / inverse orthogonal transform. The present invention relates to a video decoding device.

2. Description of the Related Art In recent years, image communication services such as videophones and video conferences are promising as services that effectively utilize ISDN (Integrated Services Digital Network) and GSTN (General Switched Telephone Network), and efficient transmission of such moving images is promising. Research on high-efficiency coding aimed at has been actively conducted. These studies use the statistical properties of an image to reduce the amount of information by removing the redundancy contained in the image. As such an encoding method, a hybrid encoding method combining motion compensation prediction and discrete cosine transform is well known.

On the decoding side, a post-filter that reduces block distortion and mosquito noise peculiar to the hybrid coding scheme that combines motion compensation prediction and discrete cosine transform on the decoded image is implemented to improve image quality. .

Hereinafter, a conventional example of a moving picture decoding apparatus using motion compensation prediction and two-dimensional orthogonal transform will be described with reference to FIG. 7, reference numeral 51 denotes a variable length decoding unit, 52 denotes an inverse quantization unit, 53 denotes an inverse orthogonal transform unit, 54 denotes an adder, 55 denotes a frame memory unit, 56 denotes a motion compensation prediction unit, 57 denotes a switch, and 58 denotes a post. 3 illustrates a filter unit.

(4) The coded bit stream that is the coded information is input to the variable length decoding unit 51. The coded bit stream includes an orthogonal transform coefficient of an image generated by the moving image coding apparatus and a coding mode indicating a coding state, that is, whether or not to perform coding, a motion vector, information on inter-frame / intra-frame prediction, and the like. Are variable-length coded and multiplexed.

The variable length decoding unit 51 decodes the coded bit stream to obtain quantized orthogonal transform coefficients, motion vectors, and coding mode information such as inter-frame / intra-frame prediction. The inverse quantization unit 52 performs inverse quantization on the quantized orthogonal transform coefficient to obtain an orthogonal transform coefficient, and the inverse orthogonal transform unit 53 performs two-dimensional inverse orthogonal transform to obtain a difference image.

When the encoding mode is the inter-frame prediction, the output of the switch 57 becomes the output of the motion compensation prediction unit 56, and the difference image is decoded by the adder 54 into the previous frame stored in the frame memory unit 55. The decoded image of the current frame is accumulated in the frame memory unit 55 by adding the motion compensation prediction value from the motion compensation prediction unit 56 based on the image.

If the block is an intra-frame prediction, the switch 57 is disconnected from the motion compensation prediction unit 56 and there is nothing to be added to the output from the inverse orthogonal transform unit 53. Are stored as decoded images in the frame memory unit 55 as they are.

The post-filter unit 58 applies a post-filter for reducing block distortion and mosquito noise to the decoded image of the current frame output from the adder 54, and outputs a decoded image with reduced noise.

In the conventional method described above, the post-filter is always performed on the decoded image, and the filtering is also performed on a portion that does not require the post-filter processing, such as a still image portion that does not change from the previous frame. There are disadvantages such as waste of processing time. This leads to an increase in power consumption when hardware is used.

Also, since the coding attribute of the block is not referred to, it is not possible to perform filtering suitable for the generated noise, so that there is a disadvantage that not only the processing amount is increased but also the image quality is deteriorated due to the filtering.

In view of such problems, the present invention provides a block distortion reduction filter and a mosquito noise reduction filter based on the presence / absence of coding of a current block, a motion vector, information on inter-frame / intra-frame prediction, and a state of a transform coefficient. On / off control, and furthermore, adaptively perform block replacement by a decoded image subjected to post-filtering of the previous frame to obtain a decoded image with reduced noise due to the post-filter while reducing the processing amount. And a video decoding device capable of improving image quality.

According to the present invention, a filter means for decoding an image coded for each block, changing a pixel value sandwiching the block boundary to smoothly connect the block boundary of the decoded image, and setting a pixel value after filtering, A frame memory unit for storing the decoded image after filtering, storing the decoded image after filtering in the frame memory unit as a previous frame, and obtaining a decoded image of the current frame by extracting a block from the decoded image after filtering. The above problem is solved by reading and outputting.

According to the present invention, a decoded image after filtering that smoothly connects block boundaries is stored as a decoded image of the previous frame, and the decoded image of the previous frame is output as a block of the current frame based on the state of the transform coefficient and the motion vector. Therefore, it is possible to obtain a decoded image with reduced block distortion while reducing the amount of processing by filtering. Further, when the embodiment according to the present invention is implemented as hardware, the reduction of the processing amount has the effect of reducing the power consumption.

Hereinafter, an embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram of an embodiment of the present invention. A variable length decoding unit 1 for performing variable length decoding of an encoded bit stream, and a quantized orthogonal transform coefficient connected to the variable length decoding unit 1 and subjected to inverse quantization. An inverse quantization unit 2 connected to the inverse quantization unit 2, an inverse orthogonal transformation unit 3 connected to the inverse quantization unit 2 to obtain a difference image by performing a two-dimensional inverse orthogonal transformation on the orthogonal transformation coefficient, and an inverse orthogonal transformation unit 3 and a switch 7. An adder 4 for connecting the difference image obtained by the inverse orthogonal transform unit 3 and the motion compensation prediction value to obtain a decoded image; and a decoded image connected to the adder 4 and obtained by the adder 4. A first frame memory unit 5 for accumulating the data, a variable length decoding unit 1, a motion compensation prediction unit 6 connected to the first frame memory unit 5 for outputting a motion compensation prediction value, and a variable length decoding unit 1. A switch 7 connected to the motion compensation prediction unit 6 to switch the output of the motion compensation prediction value; A post-filter processing content control unit 8 connected to the decoding unit 1 and controlling the operation of the post-filter; a post-filter unit 9 connected to the adder 4 and the post-filter processing content control unit 8 to perform a post-filter on the decoded image A second frame memory unit 10 connected to the post-filter processing content control unit 8 and the output signal selection unit 11 for storing the post-filtered image; the post-filter unit 9 and the second frame memory unit 10 and an output signal selection unit 11 connected to the post-filter processing content control unit 8 for switching and controlling the output of the post-filtered image.

動作 The operation of the video decoding device having the above configuration is as follows.

The variable-length decoding unit 1 performs a DCT on an image by the moving image encoding apparatus, quantizes the DCT coefficient, quantizes the DCT coefficient, information on whether or not encoding is performed, intra-frame / inter-frame prediction. A coded bit stream generated by performing variable-length coding on a prediction coding mode, a motion vector, and the like, which is information of the ITU-T Recommendation H.264. An H.263 encoded bit stream is input.

Then, by performing variable-length decoding here, discrete cosine transform (DCT) coefficients quantized for each block, motion vectors, information on inter-frame / intra-frame prediction (prediction coding mode), blocks not to be coded ( Information on whether or not encoding has been performed), and encoded information such as the quantization step size are obtained.

The inverse quantization unit 2 inversely quantizes the quantized DCT coefficient obtained by the variable length decoding unit 1 according to the quantization step size to obtain an inversely quantized DCT coefficient.

The inverse orthogonal transform unit 3 performs two-dimensional inverse DCT on the inversely quantized DCT coefficient to obtain a difference image.

When the block is inter-frame prediction, the adder 4 calculates a motion compensation prediction value calculated based on the motion vector of the block from the decoded image of the previous frame stored in the first frame memory unit 5 or the block. If the prediction is intra-frame prediction, the switch 7 switches the output so that the motion compensation prediction value becomes zero, and adds the difference image obtained by the inverse orthogonal transform unit 3 to obtain a decoded image.

The first frame memory unit 5 stores the decoded image obtained by the adder 4. The motion compensation prediction unit 6 obtains a motion compensation prediction value from the decoded image of the previous frame in the first frame memory unit 5 and the motion vector as information from the variable length decoding unit 1.

The switch 7 switches the input to the adder 4 according to the information on the prediction coding mode decoded by the variable length decoding unit 1.

The output from the adder 4 is output to the first frame memory 5 and also to the post-filter 9.

The post-filter processing content control unit 8 controls the operations of the post-filter unit 9, the second frame memory unit 10, and the output signal selection unit 11 so as to reduce the processing amount and perform optimal noise reduction processing. Do. As shown in FIG. 2, the following processing is determined according to the coding mode of the current block, which is the coding information output from the variable length decoding unit 1, and the state of the DCT coefficient.

The encoding mode can be easily determined by the bit indicating the encoding mode existing in the bit stream. As to the means for determining the state (presence / content) of the DCT coefficient, if a bit flag indicating the state of the DCT coefficient exists in the bit stream, the determination may be made based on the bit flag. If such a bit flag does not exist in the bit stream, the state of the DCT coefficient of each block is determined based on the variable-length decoded data itself.

(1) When the block is not coded The position of the current block of the decoded image in which neither the block distortion reduction filter nor the mosquito noise reduction filter is implemented and the post-filter of the previous frame in the second frame memory unit 10 is implemented Is output.

(2) When no DCT coefficient exists in the block in the inter-frame prediction Neither the block distortion reduction filter nor the mosquito noise reduction filter is performed, and the post-filtering of the previous frame in the second frame memory unit 10 is performed. An image at a position where motion compensation has been performed on the image according to the motion vector is output.

(3) When only the DC component of the DCT coefficient exists in the block in the inter-frame prediction Only the block distortion reduction filter is applied to the output from the adder 4, and this image is output.

(4) When DC and AC components of DCT coefficients exist in a block in inter-frame prediction Both the block distortion reduction filter and the mosquito noise reduction filter are performed on the output from the adder 4.

(5) When only the DC component of the DCT coefficient exists in the block in intra-frame prediction Only the block distortion reduction filter is applied to the output from the adder 4 with a stronger filter strength than that in the case of inter-frame prediction. Then, this image is output.

(6) When DC and AC components of DCT coefficients are present in a block in intra-frame prediction A block distortion reduction filter and a mosquito for the output from the adder 4 with a stronger filter strength than in the case of inter-frame prediction Both noise reduction filters are implemented, and this image is output.

(4) The reasons for the six cases are as follows. First, if the current block is a non-coded block, it is the same as the post-filtered decoded image at the same position as the current block in the previous frame in the second frame memory unit 10. Therefore, if the current block is replaced with the current block, it is not necessary to newly perform a post filter, and the processing amount is reduced.

If the current block has no DCT coefficient in the inter-frame prediction in the inter-frame prediction, the position of the current block in the previous frame in the second frame memory unit 10 is post-filtered at the position where the motion is compensated by the motion vector. This is the same as the decoded image. Therefore, if the current block is replaced with the current block, it is not necessary to newly perform a post filter, and the processing amount is reduced.

Also, when only the DC component of the DCT coefficient exists in the block in the inter-frame prediction of the current block, only the block distortion is generated, and therefore, only the block distortion reduction filter may be performed. Therefore, the processing amount is reduced by not performing the mosquito noise reduction filter.

If the current block has both DC and AC components of DCT coefficients in the inter-frame prediction, both the block distortion and the mosquito noise may have occurred. It is necessary to implement both noise reduction filters.

Also, when only the DC component of the DCT coefficient exists in the block in the current frame in intra-frame prediction, only block distortion occurs, and therefore only the block distortion reduction filter needs to be implemented. Therefore, the processing amount is reduced by not performing the mosquito noise reduction filter. However, the intra-frame prediction is more highly compressed than the inter-frame prediction, so that the image quality is degraded. Therefore, the filtering process is performed with a stronger filter strength than in the inter-frame prediction.

If the current block has intra-frame prediction and DC and AC components of DCT coefficients are present in the block, it is possible that both block distortion and mosquito noise have occurred. Both reduction filters need to be implemented. However, the intra-frame prediction is more highly compressed than the inter-frame prediction, so that the image quality is degraded. Therefore, the filtering process is performed with a stronger filter strength than in the inter-frame prediction.

の 一 One example of the above processing will be described with reference to FIG. Originally, the decoded image is output from the adder 4 in block units. However, in FIG. 3A, the output content of the adder 4 is shown in a form in which blocks are arranged in the same manner as the restored image for the sake of explanation. . The content of the second frame memory in FIG. 3B is an image on which the post-filter of the previous frame has been performed.

に お い て In FIG. 3, since block 1 is exactly the same as the previous frame, the information on whether or not coding has been performed is a block that is not coded. Therefore, it is sufficient to use the block indicated by the block 2 of the image on which the post-filter of the previous frame has been executed, which is stored in the second frame memory. There is no.

{Circle around (3)} Since block 3 is the same as block 4 indicated by a motion vector, it has a motion vector in inter-frame prediction and has no DCT coefficient. Therefore, it is sufficient to use the block indicated by block 4, and there is no need to perform block distortion and mosquito noise reduction processing.

{Circle around (5)} Since the block 5 is an image that does not exist at all in the previous frame, DC and AC components of DCT coefficients exist in the block in intra-frame prediction. Therefore, block distortion and mosquito noise reduction processing are performed on the output from the adder 4. The post-filter unit 9 reduces noise generated in the decoded image in the adder 4.

FIG. 4 is a diagram for explaining the configuration of the post-filter unit 9. The post-filter processing content control unit 8 is connected to the block filter unit 22, the switching unit 23, the mosquito noise reduction filter unit 24, and the switching unit 25. A filter switching control section 21 for controlling the operation of the adder 4, a block distortion reduction filter section 22 connected to the adder 4 and the filter switching control section 21 for performing a filter for reducing block distortion, the adder 4, and the filter switching. A switching unit 23 connected to the control unit 21 and the block distortion reduction filter unit 22 for switching and outputting the output of the block distortion reduction filter or the output of the adder 4; and a connection to the filter switching control unit 21 and the switching unit 23. A mosquito reduction filter section 24 for performing a filter for reducing mosquito noise; a filter switching control section 21; Skeet noise reduction or connect outputs of mosquito noise reduction filter in the filter unit 24, constituted by switcher 25 to output switching whether the output of the switching unit 23.

The filter switching control unit 21 controls the post-filtering operation as follows in accordance with the processing determined by the post-filter processing content control unit 8.

(1) When neither the block distortion reduction filter nor the mosquito noise reduction filter is executed The block distortion reduction filter unit 22 is turned off, and the output of the adder 4 becomes the output of the switch 23 as it is. Further, the mosquito noise reduction filter unit 24 is turned off so that the output of the switch 23 becomes the output of the switch 25 as it is.

(2) When Only the Block Distortion Reduction Filter is Executed The block distortion reduction filter unit 22 is turned on so that the output of the block distortion reduction filter unit 22 becomes the output of the switch 23. Further, the mosquito noise reduction filter unit 24 is turned off so that the output of the switch 23 becomes the output of the switch 25 as it is.

(3) When Both Block Distortion Reduction Filter and Mosquito Noise Reduction Filter are Implemented The block distortion reduction filter unit 22 is turned on so that the output of the block distortion reduction filter unit 22 becomes the output of the switch 23. Further, the mosquito noise reduction filter section 24 is turned on so that the output of the mosquito noise reduction filter section 24 becomes the output of the switch 25.

The block distortion reduction filter unit 22 performs the following processing when executing the block distortion reduction filter according to the control of the filter switching control unit 21. Since the purpose of this is to remove the discontinuity at the block boundary, two pixels x0 and x1 sandwiching the block boundary of the decoded image, which is the output of the adder 4, are smoothly connected as shown in FIG. To do. At this time, in order to prevent overcompensation of a portion having a large level difference, nonlinear processing such as Expression (1) is performed, and x0 is replaced with an output pixel value y0. Here, q is a parameter for controlling the filter strength.

{Circle around (4)} When the control of the filter switching control unit 21 turns off the block distortion reduction filter unit 22, the above processing is not performed.

When the control of the filter switching control unit 21 turns on the block distortion reduction filter unit 22, the switching unit 23 switches to output the output of the block distortion reduction filter unit 22, that is, the decoded image with reduced block distortion. Is performed. When the control of the filter switching control unit 21 turns off the block distortion reduction filter unit 22, switching is performed so that the output of the adder 4, that is, the decoded image that has not been subjected to the filtering process is output as it is.

The mosquito noise reduction filter unit 24 performs the following processing when implementing the mosquito noise reduction filter under the control of the filter switching control unit 21. Here, as shown in FIG. 6, an output pixel yi, j is obtained by using a filter shown in Expression (2) using a 3 × 3 pixel window centered on the target pixel xi, j, and is replaced with xi, j. Here, ε is a parameter for controlling the filter strength.

{Circle around (4)} When the control of the filter switching control unit 21 turns off the mosquito noise reduction filter unit 24, the above processing is not performed.

When the control of the filter switching control unit 21 turns on the mosquito noise reduction filter unit 24, the switching unit 25 switches the output of the mosquito noise reduction filter unit 24, that is, the decoded image in which the mosquito noise is reduced. Is performed. When the control of the filter switching control unit 21 turns off the mosquito noise reduction filter unit 24, the output of the switching unit 23, that is, the decoded image that has not been subjected to the filter processing, or the decoded image that has been subjected to only the block distortion reduction processing, is output. The switching is performed as follows.

The second frame memory unit 10 receives the output of the output signal selection unit 11 and stores it as a decoded image of the previous frame in which noise is reduced. Also, under the control of the post-filter processing content control unit 8, if the current block is a block that is not coded, the image of the previous frame corresponding to that position is input to the output signal selection unit 11 in block units. When no DCT coefficient exists in the block in the inter-frame prediction, the image at the position where the motion compensation has been performed in accordance with the motion vector is input to the output signal selection unit 11.

The output signal selection unit 11 switches so as to output either the output from the post-filter unit 9 or the output from the second frame memory unit 10 according to the control of the post-filter processing content control unit 8. . When the current block is a non-coded block, or when there is no DCT coefficient in the block in the inter-frame prediction, switching is performed so that the input from the second frame memory unit 10 is output. Also, when only the DC component of the DCT coefficient exists in the block in the inter-frame prediction, or when the DC and AC components of the DCT coefficient exist in the block in the inter-frame prediction, or when the DCT coefficient exists in the block in the intra-frame prediction. When only the DC component exists, or when the DC and AC components of the DCT coefficient exist in the block in intra-frame prediction, switching is performed so that the input from the post-filter unit 9 is output.

As described above, the post-filter and the second frame memory in which the image of the previous frame subjected to the post-filter is stored are adaptively controlled based on whether or not the coding is performed, the motion vector, and the information on the inter-frame / intra-frame prediction. By doing so, it is possible to obtain a decoded image with reduced noise while reducing the processing amount.

That is, the decoded image after filtering that connects the block boundaries smoothly is stored as the decoded image of the previous frame, and the decoded image of the previous frame is output as a block of the current frame based on the transform coefficients and the state of the motion vector. Thus, it is possible to obtain a decoded image with reduced block distortion while reducing the processing amount due to. Further, when the present embodiment is implemented by hardware, the reduction of the processing amount has the effect of reducing the power consumption.

It is a block diagram showing composition of one embodiment in the present invention. It is an explanatory view showing processing of a post filter processing contents control part in the present invention. FIG. 4 is a diagram illustrating an example of a post-filter process according to the present invention. FIG. 3 is a diagram illustrating details of a configuration of a post filter unit according to the present invention. FIG. 3 is a diagram illustrating a configuration of a block distortion reduction filter according to the present invention. FIG. 3 is a diagram illustrating a configuration of a mosquito noise reduction filter according to the present invention. FIG. 25 is a block diagram illustrating a configuration of a conventional video decoding device.

Explanation of reference numerals

DESCRIPTION OF SYMBOLS 1 ... Variable length decoding part, 2 ... Inverse quantization part, 3 ... Inverse orthogonal transformation part, 4 ... Adder, 5 ... First frame memory part, 6 ... Motion compensation prediction part, 7 ... Switch, 8 ... Post filter Processing content control unit, 9 post filter unit, 10 second frame memory unit, 11 output signal selection unit

Claims (3)

Filter means for decoding an image encoded for each block, changing a pixel value sandwiching the block boundary to smoothly connect the block boundary of the decoded image, and setting a pixel value after filtering,
Comprising frame memory means for storing the decoded image after filtering,
A moving picture decoding apparatus, wherein a decoded picture after the filtering is stored as a previous frame in the frame memory means, and a block is read out from the filtered decoded picture and output when obtaining a decoded picture of a current frame. When the encoded data of the block does not include a transform coefficient and does not include a motion vector, a block at the same position as the block in the previous frame stored in the frame memory is output as a block of the current frame. 2. The video decoding device according to claim 1, wherein: When the coded data of the block does not include a transform coefficient and includes a motion vector, a block at a position where motion is compensated by the motion vector with respect to the position of the block in a previous frame stored in the frame memory. 2. The moving picture decoding apparatus according to claim 1, wherein the moving picture is outputted as a block of a current frame.

Images