KR100569388B1 - Moving picture quantization apparatus and method - Google Patents

Abstract

The present invention enables to reduce the quantization complexity and the size of the semiconductor chip for quantization by way of selectively designing and supporting only some of the prescribed quantization parameters. Unlike the conventional method of designing the chip to support the parameter to support all the quantization parameters, the chip is designed to selectively support only the quantization parameters mainly used among a plurality of defined quantization parameters, and the unsupported quantization parameters In this case, the prescribed quantization parameter nearest to the smaller side is determined as an approximate quantization parameter and the quantization DCT coefficient of the corresponding macroblock can be used to reduce the complexity of the quantization and the chip size used for quantization. It can be.

Description

Video quantization device and method {MOVING PICTURE QUANTIZATION APPARATUS AND METHOD}

1 is a schematic block diagram of a typical video encoding system suitable for applying a video quantization apparatus according to the present invention;

2 is a block diagram of a video quantization apparatus according to a preferred embodiment of the present invention.

<Description of the code | symbol about the principal part of drawing>

202: quantization parameter support decision block

204: Quantizer

206: Approximate Quantization Parameter Determination Block

208: Approximate Quantizer

210: Approximate quantization coefficient calculation block

The present invention relates to a technique for encoding video, and more particularly, to a video quantization apparatus and method suitable for quantizing DCT coefficients in video encoding using MC-DCT.

As is well known, most hybrid coding techniques use motion compensated DCPM (differential pulse code modulation), two-dimensional DCT (discrete cosine transform), quantization of DCT coefficients, VLC (variable modulation coding), and the like. The motion compensation DPCM determines a motion of an object between a current frame and a previous frame and predicts the current frame according to the motion of the object to generate a difference signal representing the difference between the current frame and the predicted frame.

In general, two-dimensional DCT uses or eliminates spatial redundancy between image data, and converts digital image data blocks, such as 8 X 8 blocks, into DCT coefficients, which are quantized, zigzag scan, VLC, etc. Through processing, the amount of data to be transmitted can be effectively reduced (or compressed).

Therefore, an image bit stream encoded through the encoding scheme as described above, that is, the encoding scheme such as motion compensation DPCM, two-dimensional DCT, DCT coefficient quantization, and VLC (or entropy encoding) is provided at the output side of the video encoding system. The next transmission point stored in the transmission buffer is sent to the transmitter for transmission to the remote receiver. At this time, the transmission time point here is related to the size (ie, capacity) and transmission rate of the transmission buffer, and is controlled so that a malfunction (data overflow or data underflow) does not occur in the transmission buffer.

More specifically, the amount of bits generated for each frame at the time of encoding is changed due to various factors (for example, the complexity of the image). In view of this, in the video encoding system, the average bit rate may be kept constant. Control of the output buffer. That is, the video encoding system checks the bit generation amount up to the frame currently encoded based on the data fullness state information of the output transmission buffer, and adjusts the bit amount to be allocated in the current frame.

In other words, in the conventional typical video encoding system, the amount of bits generated in the encoding system is controlled by adjusting the quantization parameter, that is, the quantization step size (QP), based on the data fullness state information of the output transmission buffer. If the generation amount is large, the bit generation amount is controlled by largely adjusting the quantization step size to reduce the bit generation amount, and in the opposite case, the bit generation amount is controlled through the method of increasing the bit generation amount by adjusting the quantization step size smaller.

To this end, in order to quantize the DCT coefficients, a semiconductor chip is designed to support quantization for all the quantization parameters Qp defined for quantization. In this case, the complexity and size of the quantization are increased. This problem, after all, is a factor that increases the manufacturing cost of the semiconductor chip for quantization.

Accordingly, the present invention is to solve the above problems of the prior art, it is possible to reduce the quantization complexity and the size of the semiconductor chip for the quantization through a method of selectively designing and supporting only a part of a plurality of prescribed quantization parameters An object of the present invention is to provide a video quantization apparatus and method.

According to an aspect of the present invention, there is provided an apparatus for quantizing DCT coefficients in a system for encoding video using MC-DCT, quantization, and variable length coding, wherein a quantization parameter is determined based on a quantization control signal. And means for determining whether the determined quantization parameter belongs to a quantization supportable parameter among a plurality of preset quantization parameters, and when the determined quantization parameter belongs to the quantization support parameter, an input DCT using the determined quantization parameter Means for quantizing coefficients to generate quantization coefficients, means for determining an adjacent quantization support parameter as an approximate quantization parameter when the determined quantization parameter does not belong to the quantization support parameter, and an input DCT using the determined approximate quantization parameter.Quantizing number to provide a video quantization means for generating quantized approximation coefficients.

According to another aspect of the present invention, there is provided a method of quantizing DCT coefficients in a system for encoding video using MC-DCT, quantization, and variable length encoding, wherein the quantization parameter is determined based on a quantization control signal. Determining whether the determined quantization parameter belongs to a quantization supportable parameter among a plurality of preset quantization parameters, and when the determined quantization parameter belongs to the quantization support parameter, using the determined quantization parameter Quantizing the input DCT coefficients to generate a quantization coefficient; when the determined quantization parameter does not belong to the quantization support parameter, determining a quantization support parameter close to the determined quantization parameter as an approximate quantization parameter; The present invention provides a video quantization method including approximating quantizing input DCT coefficients using an approximate quantization parameter to generate approximate quantization coefficients.

The above and other objects and various advantages of the present invention will become more apparent from the preferred embodiments of the present invention described below with reference to the accompanying drawings by those skilled in the art.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

A key technical aspect of the present invention is that, unlike the aforementioned conventional scheme of designing a chip to support all defined quantization parameters to support all quantization parameters, only the quantization parameters mainly used among the plurality of defined quantization parameters are provided. By designing the chip to support selectively, and when an unsupported quantization parameter occurs, it is possible to determine the prescribed quantization parameter adjacent to the smaller side as an approximate quantization parameter and use it to quantize the DCT coefficients of the corresponding macroblock. Through technical means, it is easy to achieve the object of the present invention.

1 is a schematic block diagram of a typical video encoding system suitable for applying a video quantization apparatus according to the present invention, and includes a first frame memory 102, a subtractor 104, a DCT block 106, and a quantization block 108. And a variable length coding (VLC) block 110, an image reconstruction block 112, an adder 114, a second frame memory 116, and a current frame prediction block 118.

Referring to FIG. 1, the current frame signal stored in the first frame memory 102 for encoding is provided to the subtractor 104 and the current frame prediction block 118 via line L11, respectively.

First, the subtractor 104 predicts the current predicted through the motion estimation and compensation provided from the current frame prediction block 218 described later through the line L13 from the current frame signal provided by the first frame memory 102 through the line L11. The frame signal is subtracted, so that the data, i.e., the error signal representing the differential pixel value, is passed to the DCT block 106, discrete cosine transformed, and then to the next quantization block 108.

Next, the quantization block 108 selects an arbitrary quantization parameter Qp (quantization step size) or an approximate quantization parameter determined according to the adaptive quantization, that is, the quantization control signal provided from the output side buffer, not shown. By quantizing the DCT coefficients by using the quantized DCT coefficients, the quantized DCT coefficients are transferred to the variable length coding block 110 and the image reconstruction block 112. Here, the quantization control signal provided to the quantization block 108 is generated according to the data fullness state information of the output side transmission buffer, and the coded bit finally generated by the quantization control signal by determining the quantization parameter Qp. The amount of generation is controlled.

As described above, a detailed process of performing adaptive quantization using the quantization parameter and the approximate quantization parameter according to the present invention through the quantization block 108 will be described in detail later with reference to FIG. 2.

Subsequently, the variable length coding block 110 variable length encodes the quantized DCT coefficients provided from the quantization block 108, and the variable length coded video signal is transferred to an output side buffer (not shown) for transmission to the receiving side. do.

Meanwhile, in the image reconstruction block 112, the quantized DCT coefficients are reconstructed into DCT and pre-quantized frame signals through inverse quantization and inverse discrete cosine transform (IDCT) and provided as an input of the adder 114. In 114), the reconstructed frame signal provided as one input and the predicted current frame signal provided from the current frame prediction block 118 described below through line L13 are added to generate a reconstructed previous frame signal. The previous frame signal is stored in the second frame memory 116. Accordingly, the immediately previous frame signal for every frame encoded through this path is continuously updated, and the reconstructed previous frame signal thus updated is the current frame prediction described below for prediction of the current frame through motion estimation and compensation. Provided to block 118.

Next, in the current frame prediction block 118, based on the current frame signal on the line L11 provided from the first frame memory 102 described above and the reconstructed previous frame signal provided from the second frame memory 116 described above. For example, motion estimation and compensation using a technique such as a block matching algorithm, that is, a predetermined block (for example, 8 ×) in a preset search range (eg, 16 × 16 search range) of a reconstructed previous frame. The current frame is predicted in units of 8 DCT blocks, and the predicted current frame signal is generated on the line L13 and provided to the subtractor 104 and the adder 114, respectively.

Therefore, in a typical video encoding system, the video is encoded at a predetermined bit rate by applying MC-DCT, protons, variable length encoding, and the like to the video through a series of encoding processes as described above.

Next, the moving picture quantization device of the present invention applicable to a typical moving picture coding system having the above-described configuration will be described.

2 is a block diagram of a video quantization apparatus according to an exemplary embodiment of the present invention, which includes a quantization parameter (Qp) support determination block 202, a quantizer 204, an approximate quantization parameter determination block 206, and an approximate quantizer. 208 and approximate quantization coefficient calculation block 210.

Referring to FIG. 2, in the quantization parameter (Qp) support determination block 202, a quantization parameter is determined based on a quantization control signal, and the determined quantization parameter is designed in a chip to support a defined quantization parameter (i.e., support for quantization). Quantization parameter).

As a result of the determination here, if the determined quantization parameter is a supported quantization parameter, the determined quantization parameter is provided to the quantizer 204. As a result, the quantizer 204 uses the determined quantization parameter in the same manner as in the conventional method. The DCT coefficients provided from the DCT block 106 of FIG. 1 are quantized, and the quantized DCT coefficients are transferred to the variable length coding block 110 and the image reconstruction block 112 of FIG. 1, respectively.

Next, if the determined quantization parameter is an unsupported quantization parameter, the quantization parameter (Qp) support determination block 202 generates a corresponding quantization parameter adjustment control signal and passes it to the approximate quantization parameter determination block 206, In response, the approximate quantization parameter determination block 206 selects the supportable quantization parameter that is closest to the determined quantization parameter (more specifically, the supportable quantization parameter that is most adjacent in the direction smaller than the determined quantization parameter). After determining Qp2), the determined approximate quantization parameter is provided to the approximated quantizer 208.

As a result, the approximate quantizer 208 quantizes the DCT coefficients provided from the DCT block 106 of FIG. 1 using the determined approximate quantization parameter Qp2, and the quantized DCT coefficients are thus approximated in the next stage. Is passed to the calculation block 210.

Next, the approximate quantization coefficient calculation block 210 calculates the final quantization coefficient, that is, the approximate quantization coefficient, using the quotient of the input DCT coefficient with the input DCT coefficient value, the quantization parameter Qp, and the approximate quantization parameter Qp2. do.

That is, assuming that the input DCT coefficient for quantization is TCO, the quantization parameter is Qp, the approximate quantization parameter is Qp2, the quotient of the DCT coefficient quantized with the approximate quantization parameter is Q2, and the final quantization coefficient is TCO '. The final quantization coefficient (approximate quantization coefficient) is obtained by the following equation.

temp = TCO-(Qp × Q2)

temp '= temp-Qp (increasing counter C until temp' is less than zero)

TCO '= Q2 + C

As an example, assuming that the input DCT coefficient is 350, the quantization parameter Qp is 31, and the approximate quantization parameter Qp2 is 30, the quotient Q2 of quantizing the input DCT coefficient with the approximate quantization parameter Qp2 is Becomes 11. Therefore, the final quantization coefficient TCO 'is calculated as shown in Equation 2 below.

temp = 350-(31 × 11) = 9

temp '= 9-31 = -22

TCO '= 11 + 0 = 11

Therefore, the approximate quantization coefficient calculation block 210 calculates the approximate quantization coefficients through a series of processes as described above, and the approximated quantization coefficients thus calculated are the variable length coding block 110 and the image reconstruction block ( 112 respectively.

As described above, according to the present invention, unlike the aforementioned conventional method of designing a chip to support all defined quantization parameters to support all quantization parameters, the quantization parameter mainly used among a plurality of defined quantization parameters By designing the chip to selectively support only the quantization parameter, and when the unsupported quantization parameter occurs, the prescribed quantization parameter closest to the smaller side is determined as an approximate quantization parameter and used to quantize the DCT coefficient of the corresponding macroblock. The complexity of quantization and chip size used for quantization can be reduced.

Claims (5)

An apparatus for quantizing DCT coefficients in a system for encoding video using MC-DCT, quantization, and variable length coding, Means for determining a quantization parameter based on the quantization control signal, and determining whether the determined quantization parameter belongs to a plurality of preset quantization parameters; Means for generating quantization coefficients by quantizing input DCT coefficients using the determined quantization parameter when the determined quantization parameter belongs to the preset quantization parameter; Means for determining an adjacent quantization parameter as an approximate quantization parameter when the determined quantization parameter does not belong to the preset quantization parameter; Means for quantizing input DCT coefficients using the determined approximate quantization parameter to generate approximate quantization coefficients Video quantization device comprising a. The method of claim 1, And the approximate quantization parameter is a preset quantization parameter that is closest in a direction smaller than the determined quantization parameter. A method of quantizing DCT coefficients in a system for encoding a video using MC-DCT, quantization, and variable length coding, Determining a quantization parameter based on the quantization control signal, Determining whether the determined quantization parameter belongs to a plurality of preset quantization parameters; Generating a quantization coefficient by quantizing input DCT coefficients using the determined quantization parameter when the determined quantization parameter belongs to the preset quantization parameter; When the determined quantization parameter does not belong to the preset quantization parameter, determining a quantization parameter close to the determined quantization parameter as an approximate quantization parameter; A process of generating approximated quantization coefficients by approximating quantizing input DCT coefficients using the determined approximate quantization parameter. Video quantization method comprising a. The method of claim 3, wherein The approximate quantization process, Obtaining a quotient by quantizing the input DCT coefficients with the determined approximate quantization parameter; Multiplying the determined quantization parameter by a quotient, subtracting a multiplication result value from the input DCT coefficient value, and calculating a remaining value; Subtracting the determined quantization parameter value from the remaining value, and subtracting the counter while increasing the counter until the result of the subtraction becomes zero or less; Adding the quotient and the counter value to determine the approximate quantization coefficient Video quantization method comprising a. The method according to claim 3 or 4, And the approximate quantization parameter is a preset quantization parameter that is closest in a direction smaller than the determined quantization parameter.

Images