JP2011049816A - Moving image encoding device, moving image decoding device, moving image encoding method, moving image decoding method, and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a moving image encoding device and a moving image decoding device for increasing frequency of applying a CBP (Coded Block Pattern), thereby improving encoding performance. <P>SOLUTION: The moving image encoding device AA is provided with: a DCT coefficient division part 61 which divides a signal of a quantized DCT coefficient (residual signal) into four in the predetermined shape between the moving image encoding device AA and the moving image decoding device BB; CBP determination parts 62A-62D which determine to perform entropy encoding for each of the quantized DCT coefficients (residual signal) divided into four; and residual signal entropy encoding parts 63A-63D which perform entropy encoding for each of the quantized DCT coefficients (residual signal) divided into four, according to determination results in each of the CBP determination parts 62A-62D. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a moving image encoding apparatus, a moving image encoding method, and a program for encoding a moving image, and a moving image decoding apparatus, a moving image decoding method, and a program for decoding a moving image encoded thereby. , Regarding.

  Conventionally, as a technique for encoding a moving image, orthogonal transform and quantization are performed on a prediction error signal (hereinafter referred to as “residual signal”), and all the DCT coefficients after quantization in a transform block are “ If “0”, 1-bit information (coded block pattern; CBP) is assigned to this block, and encoding of the residual signal in this block is omitted (for example, see Non-Patent Document 1). . This technique described in Non-Patent Document 1 is particularly effective in high compression encoding in which zero DCT coefficients are likely to occur.

  In addition, as a method for encoding moving images, encoding by orthogonal transformation and quantization is performed by allowing the expansion of the orthogonal transformation block size by using the concentration of signal energy in low-resolution components in high-resolution video. Some have been proposed to improve performance (see Non-Patent Document 2, for example). The method described in Non-Patent Document 2 is effective for high-definition video in which frequency component energy tends to gather in one place.

Joint Video Team (JVT) of ISO / IEC MPEG and ITU-VCEG, “Text of ISO / IEC 14496 10 Advanced Video Coding 3rdEdition”, July 2004. http://iphome.hhi.de/suehring/tml/index.htm

  The case where a moving image is encoded by the methods described in Non-Patent Documents 1 and 2 will be described below with reference to FIGS.

  FIG. 12 shows an N × N transform block, and it is assumed that only the block B34 includes a non-zero DCT coefficient. For this reason, when the method described in Non-Patent Document 1 is used, in the blocks B31, B32, and B33 whose DCT coefficients are all “0”, encoding of the residual signal is omitted by CBP.

  FIG. 13 shows an extension of the N × N conversion block shown in FIG. 12 to 2N × 2N. This 2N × 2N conversion block includes frequency components corresponding to the blocks B31 to B34 in FIG. For this reason, since the non-zero DCT coefficient of the block B34 in FIG. 12 is included in the 2N × 2N transform block, the entire 2N × 2N transform block is a block including the non-zero DCT coefficient. Will be treated. For this reason, all DCT coefficients corresponding to the 2N × 2N region are encoded.

  As described above, in the method described in Non-Patent Document 2, the frequency of CBP application is reduced, and the amount of code increases as the number of DCT coefficients to be encoded increases. In encoding, sufficient encoding performance may not be obtained.

  Therefore, the present invention has been made in view of the above-described problems, and an object of the present invention is to improve the coding performance by improving the application frequency of CBP.

  The present invention proposes the following matters in order to solve the above problems.

  (1) The present invention is a moving picture coding apparatus that orthogonally transforms a prediction error signal for each predetermined block size, and a prediction error signal dividing unit that divides the orthogonally transformed prediction error signal into a plurality of pieces. (For example, corresponding to the DCT coefficient dividing unit 61 in FIG. 2) and a coding determination for determining whether or not each prediction error signal divided by the prediction error signal dividing unit is to be encoded based on an orthogonal transform coefficient Encoding for encoding each prediction error signal divided by the prediction error signal dividing means in accordance with a determination result by the means (for example, corresponding to the CBP determination units 62A to 62D in FIG. 2) and the encoding determination means Means (for example, corresponding to the residual signal entropy encoding units 63A to 63D in FIG. 2) and each of the prediction error signals encoded by the encoding means to generate encoded data. Data generating means (e.g., corresponding to the DCT coefficient combination unit 64 and the encoded data generating unit 66 of FIG. 2) proposes a video encoding device, characterized in that it comprises a, a.

  According to the present invention, the moving image encoding apparatus that performs orthogonal transform on the prediction error signal for each predetermined block size includes the prediction error signal dividing unit that divides the prediction error signal after orthogonal transformation into a plurality of prediction error signals. An encoding determination unit that determines whether or not each prediction error signal divided by the error signal division unit is encoded based on an orthogonal transform coefficient, and each prediction error signal divided by the prediction error signal division unit is encoded An encoding means for encoding according to a determination result by the determination means and an encoded data generating means for generating encoded data by combining the prediction error signals encoded by the encoding means are provided. For this reason, for each prediction error signal divided into a plurality, it is possible to determine whether or not to omit encoding, and to perform encoding. Therefore, it is possible to improve the encoding performance by improving the frequency of CBP application. Can do.

  (2) In the moving image encoding apparatus according to (1), the prediction error signal dividing unit divides the prediction error signal after orthogonal transformation into frequency components corresponding to rectangular regions (for example, , Which corresponds to FIG. 8).

  According to the present invention, the prediction error signal after the orthogonal transformation is divided by the prediction error signal dividing means for each frequency component corresponding to the rectangular region. According to this, an effect similar to the effect mentioned above can be produced.

  (3) In the moving picture encoding apparatus according to (1), the prediction error signal dividing unit divides the prediction error signal after orthogonal transform into frequency components corresponding to regions corresponding to frequency bands. (For example, it corresponds to FIG. 9).

  According to the present invention, the prediction error signal after orthogonal transformation is divided by the prediction error signal dividing means for each frequency component corresponding to the region corresponding to the frequency band. According to this, an effect similar to the effect mentioned above can be produced.

  (4) The present invention relates to the moving picture encoding apparatus according to (1), wherein the prediction error signal dividing means outputs the prediction error signal after orthogonal transform for each frequency component corresponding to a region corresponding to visual importance. Has been proposed (for example, corresponding to FIG. 10).

  According to the present invention, the prediction error signal dividing means divides the prediction error signal after orthogonal transform into frequency components corresponding to regions corresponding to visual importance. According to this, an effect similar to the effect mentioned above can be produced.

  (5) The present invention relates to the moving picture encoding apparatus according to (1), wherein the prediction error signal dividing unit determines the prediction error signal after orthogonal transform according to the frequency band for each frequency component corresponding to a rectangular region. For each frequency component corresponding to each region, and for each frequency component corresponding to the visual importance, it is determined on a frame basis whether to divide the frequency component, and orthogonal transform is performed for each determined frequency component. A moving picture coding apparatus that divides a subsequent prediction error signal has been proposed.

  According to this invention, the prediction error signal dividing means converts the orthogonally transformed prediction error signal for each frequency component corresponding to the rectangular region, for each frequency component corresponding to the region corresponding to the frequency band, and visually. Of each frequency component corresponding to the area corresponding to the importance, it is determined in frame units which to divide, and the prediction error signal after orthogonal transformation is divided for each determined frequency component. For this reason, since the prediction error signal after orthogonal transformation can be divided | segmented for every suitable frequency component for every flame | frame, the application frequency of CBP can be improved further and encoding performance can be improved further.

  (6) The present invention is a video decoding device that performs orthogonal transform on a prediction error signal for each predetermined block size, and is generated in any one of (1) to (5). A moving picture decoding apparatus is provided that includes a decoded data generation unit that decodes the encoded data and generates decoded data.

  According to the present invention, a code generated in any one of the video encoding apparatuses (1) to (5) is added to the video decoding apparatus that performs orthogonal transform on the prediction error signal for each predetermined block size. There is provided a decoded data generation unit for decoding decoded data and generating decoded data. For this reason, the encoded data produced | generated in the moving image encoding apparatus in any one of (1)-(5) can be decoded.

  (7) The present invention is a moving image coding method for orthogonally transforming a prediction error signal for each predetermined block size, and a first step of dividing the orthogonally transformed prediction error signal into a plurality of steps The second step for determining whether to encode based on the orthogonal transform coefficient for each prediction error signal divided in the first step, and according to the determination result in the second step, A third step for encoding each prediction error signal divided in the first step, and a fourth step for generating encoded data by combining the prediction error signals encoded in the third step. And a moving picture coding method characterized by comprising:

  According to the present invention, the prediction error signal after orthogonal transformation is divided into a plurality of parts by the moving picture encoding method, and it is determined whether or not the divided prediction error signal is encoded based on the orthogonal transformation coefficient. It was decided. Then, according to the determination result, the divided prediction error signals are encoded, and the encoded prediction error signals are combined to generate encoded data. For this reason, for each prediction error signal divided into a plurality, it is possible to determine whether or not to omit encoding, and to perform encoding. Therefore, it is possible to improve the encoding performance by improving the frequency of CBP application. Can do.

  (8) The present invention is a moving picture decoding method for orthogonally transforming a prediction error signal for each predetermined block size, and decoding encoded data generated in the moving picture coding method of (7) And a moving picture decoding method characterized by comprising a step of generating decoded data.

  According to the present invention, the encoded data generated in the moving image encoding method of (7) is decoded to generate decoded data. Therefore, decoded data can be generated from the encoded data generated in the moving image encoding method of (7).

  (9) The present invention is a program for causing a computer to execute a moving image coding method for orthogonally transforming a prediction error signal for each predetermined block size, and the prediction error signal after orthogonal transformation is obtained. A first step of dividing into a plurality of steps, a second step of determining whether or not each prediction error signal divided in the first step is encoded based on an orthogonal transform coefficient, and the second step A third step of encoding each prediction error signal divided in the first step and a combination of the prediction error signals encoded in the third step according to the determination result in the step A program is proposed that causes a computer to execute a fourth step of generating the digitized data.

  According to the present invention, by causing a computer to execute the program, the prediction error signal after the orthogonal transformation is divided into a plurality of pieces, and whether or not each divided prediction error signal is encoded based on the orthogonal transformation coefficient is determined. It was decided to. Then, according to the determination result, the divided prediction error signals are encoded, and the encoded prediction error signals are combined to generate encoded data. For this reason, for each prediction error signal divided into a plurality, it is possible to determine whether or not to omit encoding, and to perform encoding. Therefore, it is possible to improve the encoding performance by improving the frequency of CBP application. it can.

  (10) The present invention is a program for causing a computer to execute a moving image decoding method for orthogonally transforming a prediction error signal for each predetermined block size, the moving image encoding method according to (9) The program which decodes the coding data produced | generated in 1 and makes a computer perform the step which produces | generates decoded data is proposed.

  According to the present invention, the encoded data generated by causing the computer to execute the program (9) is decoded to generate decoded data. Therefore, decoded data can be generated from the encoded data generated by causing the computer to execute the program (9).

  According to the present invention, the prediction error signal after orthogonal transformation is divided into a plurality of pieces, and for each of the plurality of prediction error signals after orthogonal transformation, it is determined whether or not to omit the coding. For this reason, the application frequency of CBP can be improved and encoding performance can be improved.

  Further, according to the present invention, an image frame can be encoded by only inter-frame coding, only intra-frame coding, or a combination of inter-picture coding and intra-frame coding. Here, still image coding in which an image frame is coded only by intra-frame coding is a subset of moving picture coding assumed in the present invention. For this reason, the present invention can be applied to still image encoding, and the encoding performance of still image encoding can be improved by applying the present invention.

It is a block diagram which shows the structure of the moving image encoder which concerns on 1st Embodiment of this invention. It is a block diagram which shows the structure of the entropy encoding part with which the said moving image encoder is provided. It is a block diagram which shows the structure of the moving image decoding apparatus which concerns on 1st Embodiment of this invention. It is a block diagram which shows the structure of the encoding data analysis part with which the said moving image decoding apparatus is provided. It is a block diagram which shows the structure of the moving image encoder which concerns on 2nd Embodiment of this invention. It is a block diagram which shows the structure of the MB unit encoding process part with which the said moving image encoder is provided. It is a block diagram which shows the structure of the entropy encoding part with which the said MB unit encoding process part is provided. It is a figure for demonstrating the division shape and division number of the quantized DCT coefficient. It is a figure for demonstrating the division shape and division number of the quantized DCT coefficient. It is a figure for demonstrating the division shape and division number of the quantized DCT coefficient. It is a block diagram which shows the structure of the encoding data analysis part with which the moving image decoding apparatus which concerns on 2nd Embodiment of this invention is provided. It is a figure for demonstrating the case where a moving image is encoded using the method which concerns on a prior art example. It is a figure for demonstrating the case where a moving image is encoded using the method which concerns on a prior art example.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. Note that the constituent elements in the following embodiments can be appropriately replaced with existing constituent elements and the like, and various variations including combinations with other existing constituent elements are possible. Accordingly, the description of the following embodiments does not limit the contents of the invention described in the claims.

<First Embodiment>
[Configuration of Moving Image Encoding Device AA]
FIG. 1 is a block diagram showing a configuration of a video encoding apparatus AA according to the first embodiment of the present invention. The moving image coding apparatus AA includes an intra-coded prediction value generation unit 1, an inter-coding prediction value generation unit 2, a mode determination control unit 3, a DCT / quantization unit 4, an IDCT / inverse quantization unit 5, and entropy coding. Unit 6, first local memory 7, and second local memory 8.

  The intra encoded prediction value generation unit 1 receives the input video a, the local decoded value c in the encoded block, and the prediction direction information d in the encoded block. The intra-coded prediction value generation unit 1 generates a prediction value based on the local decoded value c in the encoded block, and calculates a coding distortion by obtaining a difference between the generated prediction value and the input signal. To do. Then, a cost value required for encoding is calculated based on the calculated encoding distortion and the prediction direction information d in the encoded block. Then, an intra prediction value e, intra prediction information (including a motion vector) f, and an encoding cost value g are output.

  The inter-coded prediction value generation unit 2 receives the input video a, the local decoded value c in the coded block, and the prediction direction information d in the coded block, and receives the inter-predicted value h and the inter-prediction information. I (including the motion vector) and the encoding cost value j are output.

  The mode determination control unit 3 receives the encoding cost value g output from the intra encoded prediction value generation unit 1 and the encoding cost value j output from the inter encoding prediction value generation unit 2. The mode determination control unit 3 compares the input cost values g and j and selects an encoding mode suitable for the processing block.

  The DCT / quantization unit 4 receives the difference between the input video a and the prediction value e and the prediction value h corresponding to the encoding mode selected by the mode determination control unit 3. The DCT / quantization unit 4 performs DCT processing and quantization processing on the input signal, performs N × M orthogonal transformation, and outputs the result as a quantized DCT coefficient (residual signal) k.

  The IDCT / inverse quantization unit 5 receives the quantized DCT coefficient (residual signal) k as an input. The IDCT / inverse quantization unit 5 performs an inverse quantization process and an inverse DCT process on the input signal and outputs it as a pixel signal m subjected to the inverse DCT.

  The entropy encoding unit 6 inputs the quantized DCT coefficient (residual signal) k and the one corresponding to the encoding mode selected by the mode determination control unit 3 from the prediction information f and the prediction information i. And The entropy encoding unit 6 performs entropy encoding on the input signal and outputs it as encoded data b. Details of the entropy encoding unit 6 will be described later with reference to FIG.

  The first local memory 7 is a signal obtained by summing the prediction value e and the prediction value h according to the encoding mode selected by the mode determination control unit 3 and the inverse DCT pixel signal m. That is, the local decoded value c in the encoded block is input. Then, the local decoded value c in the encoded block is accumulated and appropriately supplied to the intra-coded prediction value generation unit 1 and the inter-coding prediction value generation unit 2.

  The second local memory 8 receives the prediction information f and the prediction information i corresponding to the encoding mode selected by the mode determination control unit 3, that is, the prediction direction information d in the encoded block. Then, the prediction direction information d in the encoded block is accumulated and appropriately supplied to the intra-coded prediction value generation unit 1 and the inter-coding prediction value generation unit 2.

[Configuration of Entropy Encoding Unit 6]
FIG. 2 is a block diagram showing a configuration of the entropy encoding unit 6. In the present embodiment, the entropy encoding unit 6 divides a two-dimensional orthogonal transform coefficient obtained by N × M orthogonal transform by the DCT / quantization unit 4 into four in a predetermined shape. (For example, see FIG. 8 described later). The entropy encoding unit 6 includes a DCT coefficient dividing unit 61, CBP determining units 62A, 62B, 62C, and 62D, residual signal entropy encoding units 63A, 63B, 63C, and 63D, a DCT coefficient combining unit 64, A prediction information entropy encoding unit 65 and an encoded data generation unit 66 are provided.

  The DCT coefficient dividing unit 61 receives the quantized DCT coefficient (residual signal) k in the N × M DCT block as an input. The DCT coefficient dividing unit 61 divides the quantized DCT coefficient (residual signal) k signal into four in a predetermined shape between the moving picture coding apparatus AA and the moving picture decoding apparatus BB. Then, quantized DCT coefficients (residual signals) na, nb, nc, and nd are output for each divided segment.

  The CBP determination units 62A to 62D each receive the quantized DCT coefficients (residual signals) na to nd as inputs. The CBP determination unit 62A determines whether or not all of the quantized DCT coefficients (residual signals) na in the DCT block are “0”. If it is determined that all of them are “0”, it is determined that the residual signal na is not entropy-encoded, and if it is determined that at least one is other than “0”, the residual signal na is determined. Are entropy-coded. And the flag information (CBP information) pa which shows the CBP determination result which is this determination result is output. Similarly to the CBP determination unit 62A, each of the CBP determination units 62B to 62D determines whether or not each of the residual signals nb to nd is entropy-encoded, and flag information indicating the CBP determination result that is a determination result (CBP information) Each of pb-pd is output.

  Each of the residual signal entropy encoding units 63A, 63B, 63C, and 63D receives the quantized DCT coefficients (residual signals) na to nd and the CBP information pa to pd as inputs. Based on the CBP information pa, the residual signal entropy encoding unit 63A determines whether or not the residual signal na is entropy encoded and is determined by the CBP determination unit 62A. When it is determined that the CBP determination unit 62A determines that the residual signal na is entropy-encoded, the residual signal na is entropy-coded and output to the DCT coefficient combining unit 64, and the residual signal na is entropy-coded. If it is determined that the determination has not been made, the CBP determination unit 62A outputs nothing to the DCT coefficient combining unit 64. For each of the residual signal entropy encoding units 63B to 63D, similar to the residual signal entropy encoding unit 63A, each of the residual signals nb to nd is entropy encoded in accordance with each of the CBP information pb to pd. The data is output to the DCT coefficient combiner 64.

  The DCT coefficient combining unit 64 receives outputs from the residual signal entropy encoding units 63A to 63D as inputs. The DCT coefficient combiner 64 combines the input signals and outputs the entropy-encoded data q of the quantized DCT coefficients in the N × M block.

  The prediction information entropy encoding unit 65 receives the prediction information f and the prediction information i according to the encoding mode selected by the mode determination control unit 3 and the CBP information pa to pd. The prediction information entropy encoding unit 65 conforms to the encoding mode selected from the prediction information f and the prediction information i by the mode determination control unit 3 according to the encoding syntax, and CBP information pa to pd. Are entropy-encoded and output as entropy-encoded data r relating to prediction information and CBP information.

  The encoded data generation unit 66 receives the entropy encoded data q of the N × M quantized DCT coefficients and the entropy encoded data r related to the prediction information and the CBP information. The encoded data generation unit 66 generates encoded data b according to the encoding syntax and outputs it.

[Configuration of Video Decoding Device BB]
FIG. 3 is a block diagram showing the configuration of the video decoding device BB according to the first embodiment of the present invention. The video decoding device BB includes an encoded data analysis unit 110, an intra prediction value generation unit 120, an inter prediction value generation unit 130, a prediction technique control unit 140, and a memory 150.

  The encoded data analysis unit 110 receives the encoded data b. The encoded data analysis unit 110 analyzes the content described in the encoded data b according to the encoding syntax and performs entropy decoding. Then, the entropy-decoded residual signal A and entropy-decoded prediction information B obtained as a result of entropy decoding are output. Details of the encoded data analysis unit 110 will be described later with reference to FIG.

  The prediction technique control unit 140 receives the entropy-decoded prediction information B as input. The prediction technique control unit 140 identifies whether the entropy-decoded prediction information B relates to intra prediction or inter prediction, and provides a control signal F for switching between intra prediction and inter prediction. Output.

  According to the control signal F, the intra prediction value generation unit 120 receives the entropy-decoded prediction information B and the decoded pixel value C. Specifically, in the intra prediction value generation unit 120, when the prediction method control unit 140 identifies that the entropy-decoded prediction information B relates to intra prediction, the entropy-decoded prediction information B; The decoded pixel value C is input. The intra prediction value generation unit 120 generates and outputs an intra prediction value D according to the entropy-decoded prediction information B based on the decoded pixel value C.

  According to the control signal F, the inter prediction value generation unit 130 receives the entropy-decoded prediction information B and the decoded pixel value C. Specifically, when the prediction method control unit 140 identifies that the entropy-decoded prediction information B relates to inter prediction, the inter-prediction value generation unit 130 includes the entropy-decoded prediction information B; The decoded pixel value C is input. The inter prediction value generation unit 130 generates and outputs an inter prediction value E based on the decoded pixel value C in accordance with the entropy-decoded prediction information B.

  The memory 150 receives a signal obtained by summing the entropy-decoded residual signal A and the intra prediction value D and the inter prediction value E according to the control signal F, that is, the decoded pixel value C. To do. Here, the intra prediction value D and the inter prediction value E that correspond to the control signal F are when the prediction method control unit 140 identifies that the entropy-decoded prediction information B relates to intra prediction. Is the intra prediction value D. When the prediction method control unit 140 identifies that the entropy-decoded prediction information B is related to inter prediction, it is the inter prediction value E. The memory 150 accumulates the input decoded pixel value C, and supplies it to the intra prediction value generation unit 120 or the inter prediction value generation unit 130 as appropriate when performing decoding processing on an undecoded block.

[Configuration of Encoded Data Analysis Unit 110]
FIG. 4 is a block diagram illustrating a configuration of the encoded data analysis unit 110. The encoded data analysis unit 110 includes a syntax analysis unit 111, a DCT coefficient division unit 112, a prediction information entropy decoding unit 113, a CBP information identification unit 114, and residual signal entropy decoding units 115A, 115B, 115C, and 115D. A DCT coefficient combining unit 116, and an inverse orthogonal transform / inverse quantization unit 117.

  The syntax analysis unit 111 receives the encoded data b. The syntax analysis unit 111, according to the encoding syntax, generates DCT coefficients (residual signal) G entropy-coded and quantized from the encoded data b, and 4-bit CBP information I in the N × M block. And the entropy-encoded prediction information H are extracted and output.

  The DCT coefficient dividing unit 112 receives an entropy-coded and quantized DCT coefficient (residual signal) G as an input. The DCT coefficient dividing unit 112 includes four entropy-coded and quantized DCT coefficients (residual signals) G in a shape predetermined between the moving picture coding apparatus AA and the moving picture decoding apparatus BB. And is output as entropy-coded and quantized DCT coefficients (residual signals) Ja, Jb, Jc, and Jd for each divided segment.

  The prediction information entropy decoding unit 113 receives the entropy encoded prediction information H as an input. The prediction information entropy decoding unit 113 entropy-decodes the entropy-coded prediction information H and outputs it as entropy-decoded prediction information B.

  The CBP information identification unit 114 receives the 4-bit CBP information I in the N × M block as an input. The CBP information identification unit 114 extracts CBP information in each of the four areas divided from the 4-bit CBP information I, and outputs 1-bit CBP information La, Lb, Lc, and Ld for each divided segment. To do.

  Each of the residual signal entropy decoding units 115A to 115D receives DCT coefficients (residual signals) Ja to Jd that are entropy-coded and quantized, and CBP information La to Ld, respectively. If the residual signal entropy decoding unit 115A determines that the entropy-coded and quantized DCT coefficients (residual signals) Ja are all “0”, the residual signal entropy decoding unit 115A outputs a signal whose DCT coefficients are all “0”. , And is output to the DCT coefficient combining unit 116 as an entropy decoded residual signal. On the other hand, if it is determined that at least one of the entropy-encoded and quantized DCT coefficients (residual signal) Ja is other than “0”, the entropy-encoded residual signal Ja is entropy-decoded. , And is output to the DCT coefficient combining unit 116 as an entropy decoded residual signal. For each of the residual signal entropy decoding units 115B to 115D, as in the residual signal entropy decoding unit 115A, each of the entropy-coded and quantized DCT coefficients (residual signals) Jb to Jd is “0”. It is determined whether or not, and entropy-coded and quantized DCT coefficients (residual signals) Jb to Jd are entropy-decoded according to the determination result, and the entropy-decoded residual signals are combined with DCT coefficients To the unit 116.

  The DCT coefficient combining unit 116 receives outputs from the residual signal entropy decoding units 115A to 115D as inputs. The DCT coefficient combiner 116 combines the input signals and outputs them as an entropy-decoded residual signal M in the N × M block.

  The inverse orthogonal transform / inverse quantization unit 117 receives as input the residual signal M subjected to entropy decoding in the N × M block. The inverse orthogonal transform / inverse quantization unit 117 performs inverse quantization and inverse orthogonal transform on the entropy-decoded residual signal M in the N × M block, and outputs the result as an entropy-decoded residual signal A. .

  According to the above moving picture coding apparatus AA, the quantized DCT coefficient (residual signal) k signal is shaped in a predetermined shape between the moving picture coding apparatus AA and the moving picture decoding apparatus BB. Divide into four, determine whether entropy encoding is performed for each of the four quantized DCT coefficients (residual signals) na, nb, nc, and nd, and entropy encoding according to the determination result To do. For this reason, it is possible to determine whether or not to omit encoding for each of the four quantized DCT coefficients (residual signals) na, nb, nc, and nd, and to perform encoding. Thus, the encoding performance can be improved.

  Further, according to the above moving image encoding apparatus AA, an image frame can be encoded by only inter-frame encoding, only intra-frame encoding, or a combination of inter-image encoding and intra-screen encoding. . Here, still image coding that encodes an image frame only by intra-frame coding is a subset of moving image coding assumed by the moving image coding apparatus AA. For this reason, the moving image encoding device AA can encode not only moving images but also still images, and can improve the encoding performance of still image encoding.

  Also, according to the above moving picture decoding apparatus BB, the encoded data b entropy-coded by the moving picture encoding apparatus AA is determined in advance between the moving picture encoding apparatus AA and the moving picture decoding apparatus BB. It is determined whether entropy decoding is performed for each of the four entropy-coded and quantized DCT coefficients (residual signals) Ja, Jb, Jc, and Jd. Entropy decoding according to For this reason, the encoded data b entropy-encoded by the moving image encoding device AA can be decoded.

Second Embodiment
[Configuration of Video Encoding Device CC]
FIG. 5 is a block diagram showing a configuration of a video encoding device CC according to the second embodiment of the present invention. The moving image encoding device CC controls the division shape and the number of divisions for dividing a two-dimensional orthogonal transform coefficient in units of frames. This moving image encoding device CC includes an MB unit encoding processing unit C1 and a control information entropy encoding unit C2.

  The control information entropy encoding unit C2 receives the frame unit control information t. The control information entropy encoding unit C2 entropy encodes the frame unit control information t and outputs it as frame unit control information encoded data v.

  The MB unit encoding processing unit C1 receives the input video a and the frame unit control information t. The MB unit encoding processing unit C1 encodes the input video a in MB units using the frame unit control information t, generates encoded data in each MB, and outputs the encoded data as MB unit encoded data u.

[Configuration of MB Unit Encoding Processing Unit C1]
FIG. 6 is a block diagram showing a configuration of the MB unit encoding processing unit C1. The MB unit coding processing unit C1 is different from the moving image coding apparatus AA according to the first embodiment of the present invention shown in FIG. 1 in that the frame unit control information t is input, and the entropy coding unit 6 Instead, an entropy encoding unit 6A is provided. In the MB unit encoding processing unit C1, the same components as those of the moving image encoding device AA are denoted by the same reference numerals, and the description thereof is omitted.

  In addition to the quantized DCT coefficient k and the encoding mode selected by the mode determination control unit 3 among the prediction information f and the prediction information i, the entropy encoding unit 6A performs frame unit control. Information t is taken as input. The entropy encoding unit 6A performs entropy encoding on the input signal and outputs it as MB unit encoded data u.

[Configuration of Entropy Encoding Unit 6A]
FIG. 7 is a block diagram showing a configuration of the entropy encoding unit 6A. The entropy encoding unit 6A is different from the entropy encoding unit 6 according to the first embodiment of the present invention shown in FIG. 2 in that a DCT coefficient dividing unit 61A is provided instead of the DCT coefficient dividing unit 61.

  The DCT coefficient dividing unit 61A receives the frame unit control information t in addition to the quantized DCT coefficient (residual signal) k in the N × M DCT block. The DCT coefficient dividing unit 61A divides the quantized DCT coefficient (residual signal) k signal into a division shape and the number of divisions according to the frame unit control information t, and is quantized for each divided segment. Output as a DCT coefficient (residual signal).

  The division of the quantized DCT coefficient (residual signal) k by the DCT coefficient dividing unit 61A will be described below with reference to FIGS. FIG. 8 shows a case where the quantized DCT coefficient (residual signal) k is divided into frequency components corresponding to the four square regions B1, B2, B3, and B4. FIG. 9 shows a case where the quantized DCT coefficient (residual signal) k is divided into frequency components corresponding to the regions B11, B12, B13, and B14 corresponding to the four frequency bands. FIG. 10 shows a case where the quantized DCT coefficient (residual signal) k is divided into frequency components corresponding to the regions B21, B22, B23, and B24 according to the four visual importance levels.

  The DCT coefficient dividing unit 61A selects one of the division shapes and the number of divisions shown in FIGS. 8 to 10 for each frame based on the frame unit control information t, and is quantized to the selected division shape and the number of divisions. The DCT coefficient (residual signal) k is divided.

[Configuration of Video Decoding Device DD]
The video decoding device DD according to the second embodiment of the present invention decodes the MB unit encoded data u encoded by the video encoding device CC according to the second embodiment of the present invention shown in FIG. . This video decoding device DD is different from the video decoding device BB according to the first embodiment of the present invention shown in FIG. 3 in that it includes an encoded data analysis unit 110A instead of the encoded data analysis unit 110. . In the video decoding device DD, the same components as those of the video decoding device BB are denoted by the same reference numerals, and the description thereof is omitted.

  FIG. 11 is a block diagram illustrating a configuration of the encoded data analysis unit 110A. The encoded data analysis unit 110A is different from the encoded data analysis unit 110 according to the first embodiment of the present invention shown in FIG. 4 in that a syntax analysis unit 111, a DCT coefficient division unit 112, a prediction information entropy decoding unit 113, Instead of the CBP information identifying unit 114, a syntax analyzing unit 111A, a DCT coefficient dividing unit 112A, a prediction information entropy decoding unit 113A, and a CBP information identifying unit 114A are different.

  The syntax analysis unit 111A receives the encoded data b. In accordance with the encoding syntax, the syntax analysis unit 11A generates a DCT coefficient (residual signal) G entropy-coded and quantized from the encoded data b, and entropy-coded prediction information H. At the same time, information on the CBP division shape and the number of divisions is obtained. Then, entropy-encoded and quantized DCT coefficients (residual signal) G and entropy-encoded prediction information H are output, and information on the CBP division shape and the number of divisions is used as CBP division information I Output.

  The DCT coefficient dividing unit 112A receives the CBP division information I in addition to the entropy-coded and quantized DCT coefficient (residual signal) G. The DCT coefficient dividing unit 112A converts a DCT coefficient (residual signal) G entropy-coded and quantized according to CBP division information I controlled in units of frames into a DCT coefficient dividing unit of the moving picture coding device CC. The division is performed in the same division shape and division number as the division shape and division number selected in 61A. Then, for each divided segment, entropy-coded and quantized DCT coefficients (residual signals) Ja, Jb, Jc, and Jd are output.

  The prediction information entropy decoding unit 113A receives as input the prediction information H that has been entropy-coded. The prediction information entropy decoding unit 113A entropy-decodes the entropy-coded prediction information H, obtains CBP division information K in addition to the entropy-decoded prediction information B, and the entropy-decoded prediction information B and , CBP information I is output.

  In addition to the 4-bit CBP information I in the N × M block, the CBP information identification unit 114A receives the CBP division information K as input. The CBP information identification unit 114A extracts the CBP information in each of the four areas divided from the 4-bit CBP information K, and 1-bit CBP information La based on the extracted CBP information and the CBP division information K. , Lb, Lc, and Ld. Then, CBP information La, Lb, Lc, Ld is output for each divided segment.

  According to the above moving picture coding device CC, for each frame, one of the division shape and the number of divisions shown in FIGS. 8 to 10 is selected, and the DCT coefficient (QT coefficient quantized to the selected division shape and the number of divisions ( The residual signal (k) is divided. Then, for each of the divided quantized DCT coefficients (residual signals) na, nb, nc, and nd, it is determined whether to perform entropy encoding, and entropy encoding is performed according to the determination result. For this reason, since the quantized DCT coefficient (residual signal) k can be divided into division shapes and division numbers suitable for each frame, the moving image according to the first embodiment of the present invention shown in FIG. Compared with the image coding apparatus AA, the use frequency of CBP can be further improved, and the coding performance can be further improved.

  Further, according to the above moving image encoding device CC, an image frame can be encoded only by inter-frame coding, only intra-frame coding, or a combination of inter-frame coding and intra-frame coding. . Here, still image coding that encodes an image frame only by intra-frame coding is a subset of moving image coding assumed by the moving image coding device CC. For this reason, the moving image encoding device CC can encode not only a moving image but also a still image, and can improve the encoding performance of still image encoding.

  Also, according to the above moving picture decoding device DD, the coded data b entropy-coded by the moving picture coding device CC is divided into the divided shape selected by the DCT coefficient dividing unit 61A of the moving picture coding device CC, and Divide into the same division shape and division number as the division number. Then, for each divided entropy-coded and quantized DCT coefficient (residual signal) Ja, Jb, Jc, Jd, it is determined whether or not entropy decoding is performed, and entropy decoding is performed according to the determination result. For this reason, the encoded data b entropy-encoded by the video encoding device CC can be decoded.

  The processing of the moving image encoding devices AA and CC of the present invention and the processing of the moving image decoding devices BB and DD are stored in a computer-readable recording medium, and the program recorded on the recording medium is encoded with moving images. The present invention can be realized by causing the devices AA and CC and the video decoding devices BB and DD to read and execute the devices.

  Further, the above-described program can be transmitted from the moving image encoding devices AA, CC and the moving image decoding devices BB, DD storing the program in a storage device or the like via a transmission medium or by a transmission wave in the transmission medium. May be transmitted to other computer systems. Here, the “transmission medium” for transmitting the program refers to a medium having a function of transmitting information, such as a network (communication network) such as the Internet or a communication line (communication line) such as a telephone line.

  Further, the above-described program may be for realizing a part of the above-described function. Furthermore, what can implement | achieve the above-mentioned function in combination with the program already recorded on moving image encoder AA, CC, moving image decoder BB, DD, what is called a difference file (difference program) may be sufficient.

  The embodiments of the present invention have been described in detail with reference to the drawings. However, the specific configuration is not limited to the embodiments, and includes designs and the like that do not depart from the gist of the present invention.

  For example, in each of the above-described embodiments, the quantized DCT coefficient (residual signal) is divided into four, but the number of divisions is not limited to this.

  For example, when the DCT coefficient dividing unit 61 of the moving picture coding apparatus AA performs division into three, the quantized DCT coefficients (residual signals) na, nb, and nc are output for each divided segment, and the quantum It is assumed that nothing is output as the converted DCT coefficient (residual signal) nd. In this case, the DCT coefficient dividing unit 112 of the video decoding device BB outputs CBP information La, Lb, and Lc, and outputs nothing as the CBP information Ld.

  On the other hand, when the DCT coefficient dividing unit 61 of the moving image coding apparatus AA divides the data into five, the entropy coding unit 6 is provided with a CBP determination unit 62E in addition to the CBP determination units 62A to 62D, and a residual. A residual signal entropy encoding unit 63E may be provided in addition to the signal entropy encoding units 63A to 63D. In this case, a residual signal entropy decoding unit 115E may be provided in the encoded data analysis unit 110 of the video decoding device BB in addition to the residual signal entropy decoding units 115A to 115D.

6, 6A ... entropy encoding unit 61, 61A ... DCT coefficient dividing unit 62A-62D ... CBP determining unit 63A-63D ... residual signal entropy encoding unit 64 ... DCT coefficient combining unit 65 ... Predictive information entropy encoding unit 66 ... Encoded data generation unit 110, 110A ... Encoded data analysis unit 111, 111A ... Syntax analysis unit 112, 112A ... DCT coefficient division unit 113 ... Predictive information entropy decoding unit 114, 114A ... CBP information identification unit 115A to 115D ... Residual signal entropy decoding unit 116, 116A ... DCT coefficient combining unit 117 ... Inverse orthogonal transform / inverse Quantization unit AA, CC: moving image encoding device C1: MB unit encoding processing unit BB, DD: moving image decoding device

Claims (10)

A video encoding device that performs orthogonal transform on a prediction error signal for each predetermined block size,
A prediction error signal dividing means for dividing the prediction error signal after the orthogonal transformation into a plurality of;
Encoding determination means for determining whether to encode based on orthogonal transform coefficients for each prediction error signal divided by the prediction error signal dividing means;
Encoding means for encoding each prediction error signal divided by the prediction error signal dividing means in accordance with a determination result by the encoding determination means;
Encoded data generating means for combining the prediction error signals encoded by the encoding means to generate encoded data;
A moving picture encoding apparatus comprising: The moving image encoding device according to claim 1,
The prediction error signal dividing unit divides the prediction error signal after orthogonal transform into frequency components corresponding to rectangular regions. The moving image encoding device according to claim 1,
The prediction error signal dividing unit divides the prediction error signal after orthogonal transform into frequency components corresponding to regions corresponding to frequency bands. The moving image encoding device according to claim 1,
The moving picture coding apparatus characterized in that the prediction error signal dividing means divides the prediction error signal after orthogonal transform into frequency components corresponding to regions corresponding to visual importance. The moving image encoding device according to claim 1,
The prediction error signal dividing means includes
The prediction error signal after the orthogonal transformation is obtained for each frequency component corresponding to a rectangular region, for each frequency component corresponding to a region corresponding to a frequency band, and for each frequency component corresponding to a region corresponding to visual importance. Decide which one to divide into frames,
A moving picture coding apparatus that divides a prediction error signal after orthogonal transformation for each determined frequency component. A video decoding device that performs orthogonal transform on a prediction error signal for each predetermined block size,
6. A moving picture decoding apparatus comprising: a decoded data generating unit that decodes encoded data generated in the moving picture encoding apparatus according to claim 1 and generates decoded data. A video encoding method for orthogonally transforming a prediction error signal for each predetermined block size,
A first step of dividing the prediction error signal after orthogonal transformation into a plurality of steps;
A second step of determining whether to encode based on an orthogonal transform coefficient for each prediction error signal divided in the first step;
A third step of encoding each prediction error signal divided in the first step according to the determination result in the second step;
A fourth step of generating encoded data by combining the prediction error signals encoded in the third step;
A moving picture encoding method comprising: A video decoding method for orthogonally transforming a prediction error signal for each predetermined block size,
8. A moving picture decoding method comprising the steps of decoding the encoded data generated in the moving picture encoding method according to claim 7 and generating decoded data. A program for causing a computer to execute a moving image encoding method for orthogonally transforming a prediction error signal for each predetermined block size,
A first step of dividing the prediction error signal after orthogonal transformation into a plurality of steps;
A second step of determining whether to encode based on an orthogonal transform coefficient for each prediction error signal divided in the first step;
A third step of encoding each prediction error signal divided in the first step according to the determination result in the second step;
A fourth step of generating encoded data by combining the prediction error signals encoded in the third step;
A program that causes a computer to execute. A program for causing a computer to execute a moving image decoding method for orthogonally transforming a prediction error signal for each predetermined block size,
10. A program for decoding encoded data generated by causing a computer to execute the program according to claim 9, and causing the computer to execute a step of generating decoded data.

Images