JP2629409B2 - Motion compensated prediction interframe coding device - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a motion-compensated prediction interframe coding apparatus for a television signal.

2. Description of the Related Art In recent years, with the development of video coding technology, videophones,
2. Description of the Related Art A motion-compensated predictive interframe coding apparatus has been developed as a high-efficiency coding apparatus for color moving images used in a video conference system, a CD-ROM, a digital VTR, and the like. For example, Norihiko Fuukiki, “Multidimensional Signal Processing of TV Images” (November 1988
Published on March 15th, published by Nikkan Kogyo Shimbun, Chapter 7, High Efficiency Coding,
pp213-pp291) are known.

In the motion-compensated prediction interframe encoding device, when the generated code amount is large, the prediction error or the quantization step size of the pixel value of the input television signal is increased so that video encoding can be performed at a constant frame rate. Is limited. As a conventional quantization step size determination method, C.I. C. Eye. T. TSGXV Document # 525 (CCITTSGXV Document # 525
“Title: Discription of Ref.Mode 18 (RM8), source: W
orking Party XV / 4 Specialist Group On Coding for V
There is known a motion-compensated predictive interframe coding apparatus described in "Isual Telephony, version: June. 9.1989").

Hereinafter, a conventional motion compensated prediction interframe coding apparatus will be described with reference to FIG.

In FIG. 2, reference numeral 51 denotes an input terminal to which an input television signal is input, and 53, a motion vector for calculating a motion vector of the coding block by comparing the picture signal of the coding block of the current frame and the reproduced picture signal of the previous frame. A calculating unit, 54 is an image memory unit that stores the reproduced image signals of the current frame and the previous frame,
58 is a motion compensation prediction unit that performs motion compensation prediction on the reproduced image signal of the previous frame, 60 is an inter-frame / intra-frame determination unit that determines whether a block to be encoded is to be intra-frame or inter-frame encoded, 62 Is a filter unit in a loop that performs a two-dimensional low-pass filter process on the motion compensated prediction signal,
64 is a prediction error calculation unit that calculates a prediction error by performing a difference operation between the original image signal and the prediction signal of the coding block, and 66 is a signal that is used to select a signal to be orthogonally transformed by the coding method selection signal and calculate a reproduced image. A switch unit for selecting, 68, an orthogonal transform unit for orthogonally transforming a signal to be orthogonally transformed, 70, a quantizer for quantizing orthogonal transform coefficients, 73, a quantization step size calculator for calculating a quantization step size, 74 Is a code memory unit that temporarily stores transmission frames, 76 is an inverse orthogonal transform unit that inversely orthogonally transforms the quantized orthogonal transform coefficients, 78 is a reproduced image calculation unit that calculates the reproduced image of the current frame, and 82 is a communication of the prediction error. A prediction error encoding unit for channel coding, 84 is a motion vector encoding unit for channel coding a motion vector, 86 is a multiplexer unit for forming a transmission frame from the prediction code and the motion vector code, and 89 is a transmission signal. Which is an output terminal for power.

The operation of the above configuration will be described below.

The signal is converted into a digital signal by an analog / digital conversion circuit (not shown).
The television signal divided into the line blocks is input from an input terminal 51 as an input television signal 52.

The motion vector calculation unit 53 compares the input television signal 52 with the reproduction television signal 55 of the previous frame stored in the image memory unit 54, calculates the motion of the encoded block as a motion vector, and calculates a motion vector signal 56. Output to At the same time, the motion vector calculation unit 53 uses the evaluation value at the time of calculating the motion vector to
It determines the invalidity and outputs the result to the motion vector signal 56 as a motion compensation prediction control signal. Therefore, the motion vector signal
In 56, a motion vector and a motion compensation prediction signal are superimposed.

The motion compensation prediction unit 58 performs (1) motion compensation prediction of the reproduced television signal 55 of the previous frame with a motion vector when the motion compensation prediction control signal indicates that the motion compensation prediction is valid, and (2) When the motion compensation prediction control signal indicates that the motion compensation prediction is invalid, the reproduced television signal 55 of the previous frame is output as it is as the motion compensation prediction signal 59.

The inter-frame / intra-frame determination unit 60 compares the input television signal 52 and the motion compensation prediction signal 59 in block units, determines the validity of the motion compensation prediction, and if the effectiveness of the motion compensation prediction is small, the corresponding block Is determined to be valid, and if the effectiveness of motion compensation prediction is high, it is determined that interframe coding is valid for the corresponding block, and the result is output as a coding method selection signal 61. By switching the coding method between the intra-frame coding method and the inter-frame coding method on a block-by-block basis, the following improvements can be achieved as compared with the case of coding using only the inter-frame coding method.

That is, (1) when a scene change occurs, intra-frame encoding is selected, so that image quality after the scene change can be improved. (2) When a large motion of the moving object occurs, a background area hidden behind the moving object appears. In this case, since the intra-frame encoding is selected, the image quality can be improved. Also,
In a storage media coding method used for a CD-ROM or the like, in order to realize a function of editing a reproduced image and a function of reproducing in the backward direction, a frame in which all blocks are intra-coded at a fixed frame period (this frame is referred to as “ It is necessary to insert a refresh frame, and the insertion of a refresh frame can be realized by providing the motion-compensated inter-frame coding apparatus with an intra-frame coding function.

The in-loop filter unit 62 performs a two-dimensional low-pass filter process on the coded block motion-compensated and predicted using the motion vector, and calculates a predicted signal 63. The prediction error calculation unit 64 performs a difference operation between the input television signal 52 of the coding block and the prediction signal 63, and outputs the result to the prediction error signal 65.
Output as

The switch unit 66 selects (1) the input television signal 52 as the signal 67 to be orthogonally transformed when the encoding method selection signal 61 selects the intra-frame encoding, and (2) the encoding method selection. When the signal 61 selects the inter-frame coding, the prediction error signal 65 is selected as the signal 67 to be orthogonally transformed.

The orthogonal transformation unit 68 performs orthogonal transformation on the signal 67 to be orthogonally transformed, removes high correlation between neighboring pixels of the signal 67 to be orthogonally transformed, and calculates an orthogonal transformation coefficient 69. In many cases, as the orthogonal transform method, a discrete cosine transform that has high conversion efficiency and is feasible in terms of hardware is used.

The quantization unit 70 quantizes the orthogonal transform coefficient 69 using the quantization step size 71, and calculates an orthogonal transform quantized coefficient 72.

The quantization step size calculation unit 73 calculates the quantization step size 71 from the residual code amount 75 in the code memory unit 74 by the following method.

Hereinafter, a method of calculating the quantization step size 71 in the conventional example will be described.

In this conventional example, as shown in FIG. 3, the input television signal has a size of 352 pixels in the horizontal direction and 288 lines in the vertical direction, and an area of 16 pixels in the horizontal direction and 16 lines in the vertical direction (“macro block ( Macro Block).) The quantization step size Qb is
It is calculated from equation (1) at the start of quantization in n macroblock periods.

Qb = 2 × INT [Bcont ÷ 200q] +2 (1) However, in Expression (1), it is defined as follows.

(A) INT [*] is a function that rounds down the decimal point.

Example: INT [1.5] = 1, INT [1.3] = 1, INT [1.6] = 1 (b) Bcont indicates the residual code amount of the code memory unit 74.

(C) q is a coding speed parameter, and has a relationship with the coding speed V according to the expression (2).

V = q × 64 kbit / sec (2) Example: When V = q × 64 kbit / sec, q = 1.

As is clear from the equation (1), when the residual code amount Bcont increases, the quantization step size Qb increases, the generated code amount is limited, and video signal encoding at a constant frame rate can be realized. For example, when calculating the quantization step size Qb, when the residual code amount Bcont = 700 bits, the quantization step size Qb = 8, and when the residual code amount Bcont = 6100 bits, the quantization step size Qb = 62. However,
From the first macroblock to the (n-1) th macroblock, quantization is performed with a predetermined quantization step size Qb. For example, when V = 64 kbit / sec (q = 1), Qb = 32
And

In this conventional example, the calculation cycle n of the quantization step size Qb is n
Is n = 12.

The inverse orthogonal transform unit 76 performs an inverse orthogonal transform on the orthogonal transform quantization coefficient 72, and calculates an orthogonally transformed signal 77 including a quantization error.

The switch unit 66 provides a (1) reproduced image calculation signal when the encoding method selection signal 61 selects intra-frame encoding.
The numerical value “0” signal 80 is selected as 79, and (2) the prediction signal 63 is selected as the reproduced image calculation signal 79 when the encoding method selection signal 61 selects the inter-frame encoding.

The reproduced image calculation unit 78 adds the orthogonally transformed signal 77 including the quantization error and the reproduced image calculation signal 79 to calculate the reproduced image 81 of the encoded block.

The image memory 44 stores the reproduced image signal 81 of the current frame and outputs the reproduced image signal 55 of the previous frame. The prediction error encoding unit 82 encodes the orthogonal transform quantization coefficient 72, the quantization step size 71, and the encoding method selection signal 61, and calculates a prediction error code 83. The encoding of the quantization step size 71 is
When the value of the quantization step size 71 changes, that is, only once for n macroblocks.

The motion vector encoding unit 84 encodes the motion vector signal 56 and calculates a motion vector code 85. Multiplexer unit 86
Calculates a transmission frame 87 of a predetermined format from the prediction error code 83 and the motion vector code 85.

The code memory unit 74 temporarily stores the transmission frame 87 and synchronizes with a clock signal input from outside (not shown),
It is output from an output terminal 89 as a transmission code 88. At the same time, the code memory unit 74 calculates the code amount remaining in the memory as a residual code amount 75.

However, in the above configuration, the quantization step size Qb is fixed between block periods for calculating the quantization step size (in the conventional example, between n macroblock periods). , The same quantization step size Qb between successive n blocks regardless of the characteristics of the input television signal
The quantized orthogonal transform coefficients are quantized. That is, in a continuous block belonging to the same quantization step size cycle, a block having a fine pattern to be intra-frame encoded is quantized with the same quantization step size Qb as other blocks, so that the intra-frame There is a problem that the image quality of a block having an encoded fine pattern is deteriorated.

That is, by quantizing an orthogonal transform coefficient generated from a block having a fine pattern encoded in a frame with a large quantization step size, the fineness of the original image is lost, and the “block” becomes a flat block. “Distortion” occurred and was visually recognized as a large deterioration in image quality.

On the other hand, in order to reduce the amount of code generated by encoding the quantization step size, the number of consecutive blocks quantized with the same quantization step size must be equal to or greater than a certain value (in the conventional example, n macro blocks ), The quantization step size cannot be calculated and changed for each block.

The present invention has been made in view of the above-described problems, and is a luminance signal block, and for a block to be intra-coded, continuous blocks quantized at a first quantization step size serving as a reference calculated from a generated code amount. In the block, when the average value of the pixel values is equal to or more than a certain value, a second quantization step size is calculated from the first quantization step size in proportion to the image definition of the block. It is an object of the present invention to quantize orthogonal transform coefficients using a quantization step size of 2, thereby maintaining the definition of the original image and improving the image quality as a result.

That is, for a block having an average pixel value of a luminance signal block to be encoded in a frame that is equal to or more than a certain value and having high precision, the actual quantization step size is set to be smaller than the reference quantization step size. Thus, it is possible to maintain the fineness while limiting the generated code amount, and as a result, it is possible to improve the image quality of the block.

Further, when the first quantization step size is equal to or larger than a certain value, the code amount of the luminance signal is reduced by correcting the threshold value for calculating the second quantization step size, and the code amount of the chrominance signal is reduced. Can be increased, and the color degradation of the coded block caused by the extremely small code amount of the color difference signal can be prevented. As a result, the image quality of the entire image can be improved.

Means for Solving the Problems In order to achieve this object, the present invention provides an A / D converter for converting a television signal from analog to digital, and one frame or one field of a digitized input television signal. Means for dividing a block into blocks having different sizes, a motion vector detecting means for calculating a motion vector which is a motion of a television image for each block, and determining whether motion compensation prediction is performed using a motion vector for each block. Compensation prediction means for performing motion compensation prediction on a reproduced image of a previous frame with a motion vector for a block to be compensated for motion, and calculating a prediction pixel value; and a pixel value and a prediction pixel value of an input television signal. Error calculating means for calculating a difference from the predicted value as a prediction error value; An encoding method determining unit that determines whether to perform encoding or intra-frame encoding, and a determination result of the encoding method determining unit for each block, which determines whether a signal to be orthogonally transformed is a pixel value of an input television signal or a prediction error. Switching means for switching to a value, an orthogonal transformation means for orthogonally transforming a pixel value or a prediction error value of an input television signal and calculating an orthogonal transformation coefficient, and calculating a first quantization step size from a generated code amount A first quantization step size determining means, an average and variance calculating means for calculating an average and a variance of a pixel value of each block of the luminance signal of the input television signal, and a luminance signal block, and performing intra-frame encoding. Means for correcting a predetermined threshold for classifying the blocks by average or variance by a first quantization step size; , And for intra-frame coding, the blocks are classified into classes based on the average, the variance, and the corrected threshold value, and the second quantization step size is calculated for each class from the first quantization step size. For the other blocks, a second quantization step size determining unit that sets the first quantization step size to a second quantization step size, and an orthogonal transformation coefficient using the second quantization step size are used. Quantizing means for quantizing and calculating a quantized orthogonal transform coefficient, information on intra-frame coding or inter-frame coding, information on a first quantization step size and classification of quantization, Encoding means for encoding the orthogonal transform coefficients thus obtained. Inverse quantization means for inversely orthogonally transforming the quantized orthogonal transform coefficients to calculate an inverse quantization signal; and a prediction pixel value obtained by motion compensation prediction of a pixel value used in calculating a reproduction pixel value based on a determination result of the encoding method determination means. Or a switching means for switching between a numerical value “0”, an image reproducing means for calculating a reproduced image from a predicted pixel value or a numerical value “0” and an inversely quantized signal, and an image storing means for accumulating the reproduced image. It is configured to provide a motion vector encoding means for encoding a motion vector.

Operation According to the present invention, it is considered that the fineness of each block of the input television signal can be measured by the variance σ ² of the pixel values in the block by the above configuration. For example, the variance σ ^{2 of} a block having a fine pattern is considered to be smaller than the variance σ ² of a “coarse” block having a sharp change in pixel value. Also, it is considered that the variance σ ² becomes smaller as the block has higher definition.

Also, even if the block has a small variance, when the average value of the pixel values is smaller than a certain value, it can be said that the block is hardly visually noticeable even if the definition is high.

Therefore, according to the present invention, with the above-described configuration, successive blocks which measure the fineness of each block of the luminance signal of the input television signal with the variance σ ² and quantize with the same reference first quantization step size are used. For a block having a fine image to be intra-coded and having an average pixel value equal to or larger than a certain value, the first quantization step size is set to the average and variance of the pixel value of the block and the first quantization step size. By quantizing the orthogonal transform coefficients with the second quantization step size reduced according to the threshold value corrected by the quantization step size of 1, the image quality of a block having a fine image can be reduced while limiting the amount of generated code. It is something that can be improved.

Embodiment Hereinafter, an embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram showing an embodiment of a motion compensated prediction interframe coding apparatus according to the present invention.

Reference numeral 1 denotes an input terminal to which an input television signal is input, and reference numeral 3 denotes an operation for comparing the image signal of the encoded block of the current frame with the reproduced image signal of the previous frame to calculate a motion vector of the encoded block and a motion compensation prediction control signal. A vector calculation unit, 4 is an image memory unit that stores the reproduced image signals of the current frame and the previous frame, 7 is a motion compensation prediction unit that performs motion compensation prediction on the reproduced image signal of the previous frame, and 9 is a block that encodes the block to be encoded. An inter-frame / intra-frame determination unit for determining whether to perform inter-coding or intra-frame encoding; 11, an in-loop filter unit for performing a two-dimensional low-pass filter process on a motion compensation prediction signal; 13, an original image signal of an encoded block A prediction error calculation unit that calculates a prediction error by performing a difference operation between the prediction signal and the prediction signal, and selects a signal to be orthogonally transformed and calculates a reproduced image based on the encoding method selection signal. Switch unit for selecting a signal,
17 is an orthogonal transformation unit for orthogonal transformation, 19 is a quantization unit for quantizing orthogonal transformation coefficients, 21 is a second quantization step size calculation unit for calculating a second quantization step size, and 22 is the first quantization A first quantization size calculator for calculating a step size, 25 an average / variance value calculator for calculating the average and variance of the input television signal, 28 a code memory unit for temporarily storing transmission frames, and 30 a quantized An inverse orthogonal transform unit for performing an inverse orthogonal transform of the orthogonal transform coefficient; 34, a reproduced image calculating unit for calculating a reproduced image of the current frame; 36, a coding method selection signal, a prediction error, a first quantization step size, and a quantization class A prediction error encoding unit for channel-encoding information; 38, a motion vector encoding unit for channel-coding a motion vector; 40, a multiplexer unit for forming a transmission frame from the prediction code and the motion vector code; 43, a transmission signal To Which is an output terminal for power.

The operation of the above configuration will be described below.

The television signal is analog-to-digital converted by a signal processing unit (not shown in FIG. 1), divided into blocks of M pixels in the horizontal direction and N lines in the vertical direction, and input as an input television signal 2 from the input terminal 1. .

Next, the motion vector calculation unit 3 calculates the input television signal 2
And the reproduced image 5 of the previous frame read from the image memory unit 4 to calculate a motion vector,
Output to At the same time, the motion vector calculation unit 3 determines whether the motion compensation prediction for the encoded block is valid or invalid using the evaluation value at the time of motion vector calculation, and outputs the result to the motion vector signal 6 as motion compensation prediction control information. I do.

The motion compensation prediction unit 7 performs motion compensation prediction using a motion vector when performing motion compensation prediction on the reproduced image 5 of the previous frame at the same position as the encoded block using the motion vector signal 6,
If no motion compensation prediction is performed, the motion compensation prediction signal 8 is output without any operation.

The inter-frame / intra-frame determination unit 9 compares the input television signal 2 and the motion compensation prediction signal 8 on a block-by-block basis to determine the effectiveness of the motion compensation prediction. Is determined to be valid, and if the effectiveness of motion compensation prediction is high, it is determined that interframe coding is valid for the corresponding block, and the result is output as a coding method selection signal 10. In the case of an encoding device that requires the insertion of a refresh frame, the intra-frame / inter-frame determination unit 9 outputs an encoding method selection signal 10 so that all blocks are intra-frame encoded at a fixed frame period.

The in-loop filter unit 11 performs in-loop filter processing, which is a two-dimensional low-pass filter processing, on the motion-compensated prediction signal 8 when the coding block is a block to be motion-compensated and predicted, and otherwise performs in-loop filter processing. Instead, it is output as the prediction signal 12.

The prediction error calculation unit 13 performs a difference operation between the input television signal 2 of the coding block and the prediction signal 12, and outputs the result as a prediction error signal 14.

The switch unit 15 performs (1) a signal to be orthogonally transformed when the encoding method selection signal 10 selects intra-frame encoding.
The input television signal 2 is selected as 16 and (2) the prediction error signal 14 is selected as the signal 16 to be orthogonally transformed when the encoding method selection signal 10 has selected inter-frame encoding.

The orthogonal transformation unit 17 performs orthogonal transformation on the signal 16 to be orthogonally transformed, removes high correlation between neighboring pixels of the signal 16 to be orthogonally transformed, and calculates an orthogonal transformation coefficient 18. In many cases, as the orthogonal transform method, a discrete cosine transform that has high conversion efficiency and is feasible in terms of hardware is used.

The quantization unit 19 quantizes the orthogonal transform coefficients 18 with a second quantization step size 20. Hereinafter, a method of calculating the second quantization step size 20 will be described.

(1) When the block to be coded is not a luminance signal block or is an inter-frame coded block, the second quantization step size unit 21 determines whether the first quantization step size calculation unit 22 The first quantization step size calculated from the residual code amount 23 according to the described method is output as the second quantization step size.

(2) The block to be encoded is a luminance signal block,
In the case of an intra-coded block, the average and variance of the pixel values of the coded block are calculated by the method shown below, and the corresponding block is classified based on the calculated average and variance, which is a reference for each class. A second quantization step size used in actual quantization is calculated from the first quantization step size.

The average / dispersion value calculation unit 25 calculates the average value μ and the variance σ ² of the pixel values of the input television signal 2 of the coding block according to the formulas (3) and (4), and calculates the average signal 44 and the variance signal. Output as 26. The average μ is used to determine whether to change the first step size according to the fineness of the block.

The variance σ ² has a small value in a block having high definition of the input television signal 2, and
The value is large in a block with low definition.

However, they are defined as follows in equations (3) and (4).

(A) M indicates the number of pixels in the horizontal direction of the block.

(B) N indicates the number of vertical lines in the block.

(C) P (i, j) indicates the pixel value of the address (i, j) in the block.

The second quantization step size calculation unit 21 calculates the second quantization step size 20 and the quantization class information 27 from the average signal 44, the variance signal 26, and the first quantization step size 24 of the coding block. . First quantization step size 24
Is obtained by the first quantization step size calculation unit 22 from the code remaining amount 23 in the code memory unit 28 by the method described in the above conventional example. First, the second quantization step size calculation unit 21 corrects a predetermined threshold value based on the first quantization step size. The correction method is as follows.

First quantization step size Qb and predetermined threshold
Compare thq.

(1) When Qb <= thq The thresholds th1, th2, th3, and th4 used for classification are left uncorrected.

(2) When Qb> thq The threshold values th1, th2, th3, and th4 used for classification are corrected by the following equation.

i = 1, 2, 3, 4 max is the maximum value of the quantization step size Qb. FIG. 4 shows a relationship diagram between the quantization step size Qb and the threshold values th1, th2, th3, and th4.

Next, for a block that is a luminance signal and to be intra-coded, the average signal 44, the variance signal 26, and the thresholds th1, t
h2, th3, and th4 are compared, each block is divided into four quantization classes, and the second quantization step size 20 is calculated from the first quantization step size 24 by the quantization class.
Here, the first quantization step size is Qb, and the second quantization step size is Qstep.

(1) When 0 ≦ σ ² <th1 and μ> th4 Qstep Class = 1 (2) When th1 ≦ σ ² <th2 and μ> th4 Qstep Class = 2 (3) When th2 ≦ σ ² <th3 and μ> th4 Qstep Class = 3 (4) th3 by the ≦ sigma ² cutlet mu> For th4 Qstep Class = 4 Qstep = Qb above, the luminance signal is a block, and the second quantization step size 20 of the block to be encoded frame Is smaller than the first quantization step size 24 for a block in which the average value of the pixels of the input television signal 2 is equal to or more than a certain value and the resolution is high. When the first quantization step size exceeds a certain value, the first threshold is corrected by correcting the classification threshold.
The number of blocks quantized with a quantization step size smaller than the quantization step size of the number of blocks is reduced, the code amount of the luminance signal is reduced, and the color deterioration of the coded block caused by the extremely small code amount of the color difference signal is reduced. Can be prevented.

The quantization unit 19 quantizes the orthogonal transform coefficient 14 with the second quantization step size 20, and calculates an orthogonal transform quantized coefficient 29. The inverse orthogonal transform unit 30 performs an inverse orthogonal transform on the orthogonal transform quantization coefficient 29, and calculates a signal 31 including a quantization error.

The switch unit 16 controls the (1) reproduced image calculation signal when the encoding method selection signal 10 selects intra-frame encoding.
The numerical value “0” signal 33 is selected as 32, and (2) the prediction signal 12 is selected as the reproduced image calculation signal 32 when the encoding method selection signal 10 has selected the inter-frame encoding.

The reproduced image calculation unit 34 adds the signal 31 including the quantization error and the reproduced image calculation signal 32 to generate a reproduced image 35 of the encoded block.
Is calculated. The image memory 4 stores a reproduced image signal of the current frame.
35, and outputs the reproduced image signal 5 of the previous frame.

The prediction error encoding unit 36 outputs the encoding method selection signal 10 and the first
, The quantization step size 24, the quantization class information 27, and the orthogonal transform quantization coefficient 25 are encoded, and a prediction error code 37 is calculated.

The motion vector encoding unit 38 encodes the motion vector signal 6 of the motion-compensated block and calculates a motion vector code 39.

The multiplexer unit 40 calculates a transmission frame 41 of a predetermined format from the prediction error code 37 and the motion vector code 39.

The code memory unit 28 temporarily stores the transmission frame 41, and outputs the transmission frame 41 from the output terminal 43 as a transmission code 42 in synchronization with a clock signal input from the outside (not shown). At the same time, the code memory unit 28 calculates the code amount remaining in the memory as the residual code amount 23.

As is apparent from the above description, according to the present embodiment,
At the time of encoding a luminance signal block, in a continuous block group belonging to the same quantization step size, intra-frame encoding, and in a block in which the average value of pixel values is a certain value or more, in proportion to the fineness of the block, Since the second quantization step size is calculated from the reference first quantization step size, and the orthogonal transform coefficients are quantized using the second quantization step size, the fineness of the image is not impaired. The image quality of the entire image can be improved.

Further, when the first quantization step size exceeds a certain value, the number of blocks quantized with a quantization step size smaller than the first quantization step size is reduced by correcting a threshold for classifying. Then, the code amount of the luminance signal is reduced, and as a result, the code amount of the chrominance signal increases, and it is possible to prevent the deterioration of the color of the coded block caused by the extremely small code amount of the chrominance signal.

In the above description, the variance value 26 calculated by the average / variance value calculation unit 25 is defined by the equation (3). However, if the variance can measure the definition of the input television signal 2, any other measurement scale can be used. May be. For example, since the size of a block (the number of pixels in the horizontal direction: M, the number of lines in the vertical direction: N) is generally a fixed value, there is a numerical value D shown in Expression (6) as a scale for easily calculating. Here, p (i, j) indicates the pixel value of the address (i, j) in the block, and μ indicates the average pixel value of the block.

Further, in the above description, whether to change the first quantization step size is determined by the average value of the pixel values, but may be another measure expressing the characteristics of the entire pixel value.

Further, in the above description, the quantization step size is classified into four classes, but another classification number may be used.

Further, in the above description, the first quantization step size, which is a reference, is equally divided for each classification, and the second quantization step size is determined. Another method may be used as long as it is calculated so that the conversion step size is reduced.

Further, the threshold value for classification is corrected by the equation (5), but another correction method may be used in which the threshold value decreases as the quantization step size increases.

As described above, the present invention measures the characteristics related to the fineness of the input television signal of the block to be intra-coded, and calculates the average value within a continuous block group quantized with the same quantization step size. Is greater than or equal to a certain value, and for a block to be intra-coded having a fine picture, by reducing the quantization step size, the amount of generated code is restricted, and the fineness of the original image is reduced. Since video coding can be performed without loss, it is possible to improve image quality.

In particular, the effect of improving the image quality of the refresh frame is remarkable.

[Brief description of the drawings]

FIG. 1 is a block diagram of a motion-compensated prediction interframe coding apparatus according to an embodiment of the present invention, FIG. 2 is a block diagram of a conventional motion-compensated prediction interframe coding apparatus, and FIG. FIG. 4 is a conceptual diagram showing the relationship between macroblocks, and FIG. 4 is a characteristic diagram showing the relationship with the quantization step size of the motion compensated prediction interframe coding apparatus according to one embodiment of the present invention. 1, 51 ... input terminal, 3, 53 ... motion vector calculation unit,
4, 54: image memory unit, 58: motion compensation prediction unit, 9,
60 ... intra-frame / inter-frame determination unit, 11, 62 ... intra-loop filter unit, 13, 64 ... prediction error calculation unit, 15, 66 ...
... Switch section, 17, 68... Orthogonal transform section, 19, 70... Quantization section 21,... Second quantization step size calculation section, 22, 59.
... A first quantization step size calculator, 25... An average / variance value calculator, 28, 74... A code memory unit, 30, 76... An inverse orthogonal transform unit, 34, 78. 36, 82 ... prediction error coding unit, 38, 84 ... motion vector coding unit, 41, 86
... Multiplexer, 43, 89 ... Output terminals.

Claims (1)

(57) [Claims] A / D conversion means for analog-to-digital conversion of a television signal, blocking means for dividing one frame or one field of a digitized input television signal into blocks of a predetermined size, Motion vector detecting means for calculating a motion vector which is the motion of the television image for the block, motion compensation judging means for judging whether to perform motion compensation prediction using the motion vector for each block, and a block for motion compensation prediction. Motion-compensated prediction means for performing motion-compensated prediction of the reproduced image of the previous frame with a motion vector and calculating a predicted pixel value
Error calculating means for calculating a difference between a pixel value of an input television signal and a predicted pixel value as a prediction error value, and a coding method determination for determining whether to perform inter-frame coding or intra-frame coding for each block Means, and switching means for switching between a signal to be orthogonally transformed based on the determination result of the encoding method determining means for each block as a pixel value of an input television signal or a prediction error value, and a pixel value of an input television signal or Orthogonal transform means for orthogonally transforming the prediction error value to calculate an orthogonal transform coefficient; first quantization step size determining means for calculating a first quantization step size from the generated code amount; and luminance of the input television signal. Average to calculate the average and variance of pixel values for each block of signal
Variance calculating means, means for correcting a predetermined threshold value by a first quantization step size to classify the blocks which are luminance signal blocks and to be intra-coded by average or variance, and And for blocks to be intra-coded, the blocks are classified into classes based on the average, the variance, and the corrected threshold, and the second quantization step size is determined for each class from the first quantization step size. For the other blocks, the second quantization step size determining means for setting the first quantization step size to the second quantization step size and the second quantization step size are used to calculate the orthogonal transform coefficients. And quantization means for calculating a quantized orthogonal transform coefficient, and information on intra-frame coding or inter-frame coding. Coding means for coding information relating to the quantization step size and quantization classification of 1 and quantized orthogonal transform coefficients, and inverse orthogonal transform of the quantized orthogonal transform coefficients to calculate an inverse quantized signal. A quantizing unit, a switching unit for switching between a pixel value used in calculating a reproduction pixel value and a motion compensated predicted pixel value or a numerical value “0” based on a determination result of the encoding method determining unit; Motion compensation prediction including image reproducing means for calculating a reproduced image from a value or a numerical value "0" and an inversely quantized signal, image storing means for storing the reproduced image, and motion vector coding means for coding a motion vector Interframe coding device.