JP2003125403A - Transcoder for picture code and transcoding method for picture code - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a transcoder whose operation quantity is less and which can maintain constant picture quality and to provide a transcoding method. SOLUTION: The transcoder of a picture code is provided with a decoding means 122 decoding a first picture code and obtaining the frequency components of the respective encoding blocks of a picture, an amplitude change means 123 reducing all or a part of the amplitudes of the frequency components of the encoding blocks or making them to zero so that code quantity per encoding block is reduced when the respective encoding blocks are re-encoded and a re-encoding means 125 re-encoding the frequency components of the respective blocks, where all or a part of the amplitudes of the frequency components are reduced or become zero and obtaining a second picture code. The amplitude change means low-pass-filter the frequency components of the respective blocks.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention converts an original image code obtained by encoding an original image into a reduced image code having a code total amount smaller than that of the original image code and representing an image having a quality corresponding to the code amount. The present invention relates to an image code transcoder for performing.

[0002]

2. Description of the Related Art In recent years, MPEG (Motion Picture Expert)
The s Group) method has become widespread as a method of storing or transmitting moving images and audio, and for example, CD-R (Compact
Disc-Recordable), DVD (Digital Versatile Dis)
c), digital storage media such as a hard disk (Digital
Storage Medium) is used as a recording format,
It is also used as a data coding method for digital television broadcasting. The MPEG system is also used as a data encoding system for streaming moving images and audio.

As a streaming client for streaming with streaming data composed of MPEG signals, in addition to a personal computer that receives streaming data by optical communication with a transmission bit rate of 10 Mbps, an A with a transmission bit rate of several Mbps is used.
A personal computer that receives streaming data by DSL (Asymmetric Digital Subscriber Line) and W-CDMA (Wideband-Code Division Multiple A) that receives streaming data by a data communication channel having a transmission bit rate of 384 kbps to 2 Mbps.
ccess type mobile terminals are assumed. On the other hand, the streaming server uses MPE created by the content creator.
G data is stored, but this MPEG data is
It is high quality original MPEG data. If original MPEG data based on signals of conventional television systems such as NTSC system and PAL / SECAM system is directly transmitted, a bit rate of about 6 Mbps is required.

Therefore, in order to transmit the streaming data to various streaming clients, the transcoder requires the original MPEG data, which is a secondary MPEG having a lower transmission bit rate.
Needs to be converted to data.

Conventionally, as a method of reducing the code amount by a transcoder, (1) a video signal is decoded by an MPEG decoder and then the decoded video signal is MPE
A method of encoding by a G encoder, (2) a method of reducing the code amount by adjusting the quantization scale,
(3) A method of reducing the code amount by thinning out P pictures, (4) A method of reducing the code amount by converting an I picture into a P picture, and (5) A code amount by performing downscaling. A method of reducing is proposed.

A method of decoding a video signal with an MPEG decoder and then encoding the decoded video signal with an MPEG encoder will be described.

This method is the most primitive method, in which the video signal is decoded by the MPEG decoder shown in FIG. 22, and then the video signal is encoded by the MPEG encoder shown in FIG.

Referring to FIG. 22, the MPEG decoder comprises a buffer 701, a variable length decoder 702, an inverse quantizer 703, an inverse discrete cosine transformer (IDCT) 704, an adder 705, a past image memory 706 and a future image memory. 70
7, a motion compensation predictor 708 and a switch 709. This MPEG decoder is a video stream 751
Is input and a video signal 752 is output. The inverse quantization characteristic of the inverse quantizer is determined by the quantization scale 753. The quantization characteristic will be described later. Also, MPE
The operation of each part of the G decoder is well known, and therefore its explanation is omitted.

Referring to FIG. 23, the MPEG encoder comprises a switch 801, a discrete cosine transformer (DCT) 8
02, quantizer 803, inverse quantizer 804, inverse discrete cosine transformer (IDCT) 804, adder 806, past image memory 807, future image memory 808, motion detection unit 8
09, a motion compensation predictor 810, a switch 811, a variable length encoder 813, a buffer 814, and a control unit 815. This MPEG encoder inputs the video signal 752 output from the MPEG decoder as a video signal 851 and outputs a video stream 852. Quantizer 80
3 and the quantization characteristics of the inverse quantizer are the quantization scale 853
Is determined by Since the operation of each unit of the MPEG encoder is known, its description is omitted.

In the transcoder in which the MPEG decoder and the MPEG encoder are connected in series, the buffer 8
By making the bit rate of reading the video stream from 14 different from the bit rate of the video stream 751, transcoding becomes possible. In this case, the quantization scale 853 changes according to the bit rate of reading the video stream from the buffer 814.
In addition, the GOP (Group Of
Transcoding is also possible by changing the structure of (Pictures).

A method of reducing the code amount by adjusting the quantization scale will be described.

To implement this method, for example, a transcoder as shown in FIG. 24 is used. This transcoder includes a buffer 901, a variable length encoder 902, an inverse quantizer 903, a quantizer 904, and a multiplexer 90.
5, variable length encoder 906, buffer 907, and control unit 9
08. The buffer 901, the variable length decoder 902, and the inverse quantizer 903 are the same as those of a known MPEG decoder, and therefore their explanations are omitted. Quantizer 90
4 is a transcoded quantization scale 9 output from the control unit 908 for the DCT coefficient output from the inverse quantizer 903.
Requantization according to 54. Multiplexer 905
Multiplexes the data indicating the level of the requantized DCT coefficient output from the quantizer 904 and the data other than the DCT coefficient. The variable length encoder 906 performs variable length encoding on the data output from the multiplexer 905. This variable length coding includes two-dimensional Huffman coding of DCT coefficients. The buffer 907 absorbs fluctuations in the occurrence frequency of the variable length code output by the variable length encoder 906 and outputs the transcoded video stream 952 at a constant bit rate. The control unit 908 determines the value of the transcoded quantization scale 954 according to the occupation rate of the buffer 907, a predetermined rule, and the like. The value of the transcoded quantization scale 954 can be adjusted by making the bit rate of reading the transcoded video stream 952 from the buffer 907 different from that of the original video stream 951. Also, the transcoded quantization scale 954 thus adjusted is generally different from the original quantization scale.
Therefore, the original quantization scale 953 is converted into the transcoded quantization scale 954.

Transcoded video stream 95
The MPEG decoder for decoding 2 is the MPE shown in FIG.
This is the same as the G decoder, and this MPEG decoder uses the transcoded quantization scale 954 as the quantization scale 753.

The quantizer 9 of the transcoder shown in FIG.
In 04, regardless of whether the macroblock is an intra macroblock or a non-intra macroblock, 8 × 8 pixels obtained by DCT transforming 8 × 8 pixels of each DCT block in each macroblock. Quantize the DCT coefficient of. The result of this quantization is a numerical value indicating the amplitude of the DCT coefficient after quantization. And FIG.
In the MPEG decoder shown in 2, regardless of whether the macroblock is an intra macroblock or a non-intra macroblock, the quantized DCT coefficient is restored from the numerical value indicating the amplitude of the quantized DCT coefficient and restored. The 8 × 8 DCT coefficients thus obtained are inversely DCT-transformed to restore 8 × 8 pixels.

AC coefficient F (m, n) (m, n) of the DCT coefficients of the DCT block in the intra macroblock
Is greater than or equal to 0 and an integer satisfying m + n ≠ 0) is quantized by a quantizer having no dead zone near the amplitude of zero. Since the step size S1 of this quantizer differs depending on the AC coefficient component, it can be expressed as S1 (m, n). However, the step size S1 (m, n) is S1 (m, n) = 2 × It is expressed as Intra quantization matrix (m, n) × quantization scale / 32.

Amplitude F1 (m, n) of the quantized DCT coefficient obtained by the inverse quantizer corresponding to this quantizer
Is the amplitude F of the quantized DCT coefficient output from the quantizer.
If the numerical value indicating 1 (m, n) is QF1 (m, n), then F1 (m, n) = 2 × QF1 (m, n) × intra quantization matrix (m, n) × quantization scale / 32.

The DCT coefficient F (m, n) (m, n is an integer equal to or greater than 0) of the DCT block in the non-intra macroblock is quantized by a quantizer having a dead zone near the amplitude of zero. . Since the step size S2 of this quantizer differs depending on the DCT coefficient component, S2
Can be expressed as (m, n), but the step size S2
(M, m) is expressed as S2 (m, n) = 2 × quantization scale × non-intra-quantization matrix (m, n) / 32.

Amplitude F2 (m, n) of the quantized DCT coefficient obtained by the inverse quantizer corresponding to this quantizer
Is the amplitude F of the quantized DCT coefficient output from the quantizer.
If the numerical value indicating 2 (m, n) is QF2 (m, n), then F2 (m, n) = (2 × QF2 (m, n) + k) × non-intra-quantized matrix (m, n) × Quantization scale / 32 However, k = sign (QF2 (m, n)).

Here, the quantization scale of the above two equations is q_sc
When the value of ale_type is 0, it is expressed as quantization scale = 2 × quantization scale code, and when the value of q_scale_type is 1, it is expressed as quantization scale = function (quantization scale code). Here, function shows a non-linear characteristic. FIG. 25 shows an example of the characteristics of the quantization scale code versus the quantization scale.

The q_scale_type can be set for each picture, and the quantization scale code can be set for each macroblock. Therefore, by adjusting the value of q_scale_type for each picture and adjusting the value of the quantization scale code for each macroblock so as to be large, the quantization scale for each macroblock can be increased. Then, by increasing the quantization scale, it is possible to reduce the value of the numerical value representing the DCT coefficient of the quantized DCT block and reduce the code amount.

A method of reducing the code amount by thinning out P pictures will be described.

As shown in FIG. 26, the picture can be thinned out.
The code amount can be reduced by the amount of pictures thinned out by
You can G having a structure as shown in FIG.
It is assumed that OP (Group Of Pictures) is given. Slightly
When reducing the code amount, as shown in FIG.
To B₂Picture, B_FivePicture and B₈Thin out pictures
Ku. In this case, in the decoder, for example, B₂Picture
I₁Interpolate with picture, B_FiveP the picture_FourIn the picture
Interpolate, B₈P the picture₇Just interpolate with a picture
The transcoder then re-displays the undecimated pictures.
No need to encode. Furthermore, when reducing the code amount
Is, as shown in FIG.₂Picture, B₃Piku
Cha, B_FivePicture, B₆Picture, B₈Picture and B₉
Thin out pictures. In this case, the decoder
For example, B₂Picture and B₃Picture I ₁Interpolate with picture
Then B_FivePicture and B₆P the picture_FourInterpolate with picture
Then B₈Picture and B₉P the picture₇Interpolate with picture
Transcoder is not thinned out so
There is no need to re-encode the picture. In addition, reduce the code amount
In the case of decreasing, as shown in FIG.₂Piku
Cha, B₃Picture, B_FivePicture, B₆Picture, B₈Pi
Kucha and B₉P in addition to the picture_FourThin out pictures.
In this case, in the decoder, for example, B₂Picture, B₃
Picture, P_FourPicture and B_FivePicture I₁Picture
Interpolate with, B₆Picture, B₈Picture and B ₉Picture
To P₇Complement with a picture. But P₇Picture is P
_FourSince motion-compensated prediction is originally made from the picture, decoding
In the container, P_FourIf the picture is thinned out, P₇Picture
Cannot be decrypted. So transcoder
Is P₇Picture I₁Motion-compensated prediction from the picture,
The prediction error must be encoded.

A method of reducing the code amount by converting an I frame into a P frame will be described.

The GOP is composed of I-pictures, P-pictures and B-pictures, and the code amount per picture is the largest for I-pictures, followed by P-pictures and B-pictures. Therefore, as shown in FIG. 27, the code amount is reduced by converting the I ₁₀ picture of GOP2 into the P ₁₀ picture. To do this, the transcoder must perform motion-compensated prediction of the P ₁₀ picture from the P ₇ picture and encode the prediction error.

FIG. 28 shows an example in which a method for reducing the code amount by thinning out B pictures and P pictures and a method for reducing the code amount by converting an I frame into a P frame are combined. In this example, B ₂ picture, B ₃ picture, P ₄ picture, B ₅ picture, B ₆ picture
Picture, P ₇ picture, B ₈ picture, B ₉ picture,
B ₁₁ picture, B ₁₂ picture, P ₁₃ picture, B ₁₄ picture, B ₁₅ picture, P ₁₆ picture, B ₁₇ picture, B
₁₈ picture is thinned out, I ₁₀ picture is converted into P ₁₀ picture. In this case, the P ₁₀ picture is motion-compensated and predicted from the I ₁ picture which is the closest past picture.

A method of reducing the code amount by performing downscaling will be described.

Reducing the number of pixels included in an image is called downscaling. For example, 640 (horizontal direction)
320 x 480 (vertical direction) frames (horizontal direction)
The number of pixels included in the image is reduced by reducing the frame size to × 240 (vertical direction). In this case, the DCT block before downscaling is 2 (horizontal direction) × 2
(Vertical direction) The collected one is the downscaled 1
One DCT block. The simplest downscaling is to decode the image code to obtain the decoded image and re-encode the decoded image. At the time of re-encoding, the motion vector obtained when the image code is decoded may be reused in some cases.

[0028]

By the way, streaming is performed by a large number of streaming clients at the same time. Therefore, the streaming server needs to send streaming data to many streaming clients at the same time. There are various types of streaming clients as described above, and accordingly, there are various types of bit rates of streaming data. Therefore, the streaming server must perform a large number of transcodings simultaneously in order to simultaneously generate streaming data of various bit rates.

When the primitive transcoding method is used, a large amount of processing is required since all decoding processing and all encoding processing are required.

In addition, for example, in transcoding for thinning out P pictures, transcoding for converting I pictures into P pictures, and transcoding for downscaling, a motion prediction vector is used as a macroblock for motion compensation prediction. You need each.
In some cases, the motion vector predictor before transcoding can be reused in the macroblock after transcoding, but in some cases it cannot be reused. When the motion vector predictor before transcoding cannot be reused for the macroblock after transcoding, the generation of the motion vector predictor requires a large amount of calculation, and thus such transcoding has a large amount of calculation. Therefore, doing a large number of these types of transcoding with a CPU on one streaming server is:
Considering the current CPU power, it is almost impossible. Transcoding for adjusting the quantization scale requires less computation than transcoding for thinning out P pictures, transcoding for converting I pictures into P pictures, and transcoding for downscaling. However, even if this transcoding is used, it is very difficult to perform a large number of this type of transcoding using the CPU in one streaming server in view of the current CPU capability. In particular, in the future, it is expected that the number of streaming clients performing simultaneous streaming will increase, so that this problem is expected to become serious.

Further, in transcoding in which the code amount is reduced by adjusting the quantization scale, all Ds of all DCT blocks within the same macroblock are processed.
Since the CT coefficients are uniformly quantized roughly, the image quality deteriorates. In transcoding in which the amount of code is reduced by thinning out P pictures, the frame rate drops, so the motion becomes jerky and the image quality deteriorates. In the transcoding in which the code amount is reduced by converting the I picture into the P picture, the error resistance becomes weak, so that the image quality deteriorates even if a few errors occur. In transcoding in which the amount of codes is reduced by performing downscaling, the screen becomes smaller, so the resolution drops and the image quality deteriorates.

The present invention has been made in view of the above problems, and it is an object of the present invention to provide a transcoder and a transcoding method that can maintain a constant image quality with a small amount of calculation.

[0033]

According to a first aspect of the present invention, decoding means for decoding a first image code to obtain a frequency component of each coded block of an image, and for each coded block Amplitude changing means for reducing or zeroing the amplitude of all or some of the frequency components of each coding block so that the amount of code per each coding block decreases when encoded, and all of the frequency components or Re-encoding means for re-encoding the frequency component of each block in which a part of the amplitude has decreased or has become zero, and a re-encoding means for obtaining a second image code. An image code transcoder is provided which is characterized by low-pass filtering the frequency components of a block.

The above-mentioned image code transcoder may further comprise means for making non-zero the amplitude of one or more frequency components among the frequency components whose amplitude has become zero by the amplitude changing means.

The above-mentioned image code transcoder measures means for measuring the reduction rate from the amount of the first image code to the amount of the second image code, and the measured reduction rate is the target reduction rate. So that the frequency characteristic of the low-pass filter is changed.

According to a second aspect of the present invention, decoding means for decoding the first image code to obtain the frequency component of each coded block of the image, and each decoding block when the coded block is re-coded To reduce the amount of code per coding block,
Amplitude changing means that reduces or reduces the amplitude of all or part of the frequency components of each encoded block, and re-creates the frequency components of each block in which the amplitude of all or part of the frequency components is reduced or becomes zero. In a transcoder for an image code, which comprises a re-encoding means for encoding to obtain a second image code, the amplitude changing means divides all or some of the frequency components by an integer greater than 1, and calculates the quotient. An image code transcoder is provided, which is rounded down to an integer and the result of the integer is multiplied by the integer.

The above-mentioned image code transcoder may further comprise means for making non-zero the amplitude of one or more frequency components among the frequency components whose amplitude has become zero by the amplitude changing means.

The above-mentioned image code transcoder measures the reduction rate from the amount of the first image code to the amount of the second image code, and the measured reduction rate is the target reduction rate. The control means for changing the integer may further be provided.

According to a third aspect of the present invention, a decoding means for decoding the first image code to obtain the frequency component of each coded block of the image, and each decoding block when the coded block is re-coded To reduce the amount of code per coding block,
Amplitude changing means that reduces or reduces the amplitude of all or part of the frequency components of each encoded block, and re-creates the frequency components of each block in which the amplitude of all or part of the frequency components is reduced or becomes zero. In a transcoder for an image code, which comprises re-encoding means for encoding to obtain a second image code, the amplitude changing means includes a non-zero frequency component and the non-zero frequency component when the frequency component of each encoded block is scanned. Of the non-zero frequency component so that the length of the variable-length code corresponding to the number of zeros preceding the frequency component is reduced by one or several steps. A transcoder is provided.

In the above-mentioned image code transcoder, a variable-length code corresponding to the number of zeros leading to the non-zero frequency component and the non-zero frequency component, the non-zero frequency component of which amplitude is reduced or made zero. May be limited to those in which the reduction ratio of the amplitude after the length is reduced to be shorter than the amplitude before the reduction is less than or equal to a predetermined ratio.

In the above-mentioned image code transcoder, a non-zero frequency component whose amplitude is reduced or reduced to zero corresponds to the non-zero frequency component and the number of zeros preceding the non-zero frequency component. May be limited to those in which the difference between the amplitude after being reduced to a shorter length and the amplitude before being so reduced is less than or equal to a predetermined value.

The above-mentioned image code transcoder may further comprise means for making the amplitude of one or more frequency components among the frequency components whose amplitude has become zero by the amplitude changing means non-zero.

The above-mentioned image code transcoder measures the reduction rate from the amount of the first image code to the amount of the second image code, and the measured reduction rate is the target reduction rate. Therefore, the control means for changing the number of frequency components for decreasing or zeroing the amplitude may further be provided.

The above-mentioned image code transcoder measures the reduction rate from the amount of the first image code to the amount of the second image code, and the measured reduction rate is the target reduction rate. So as to reduce the amplitude or change the degree to which the amplitude is reduced to zero.

According to a fourth aspect of the present invention, the decoding means for decoding the first image code to obtain the frequency component of each coded block of the image and the decoding means for each coded block when re-coded To reduce the amount of code per coding block,
Amplitude changing means that reduces or reduces the amplitude of all or part of the frequency components of each encoded block, and re-creates the frequency components of each block in which the amplitude of all or part of the frequency components is reduced or becomes zero. In a transcoder for an image code, which comprises a re-encoding means for encoding to obtain a second image code, the amplitude changing means is one or more from the end when the frequency component of each encoded block is scanned. An image code transcoder is provided which is characterized by zeroing out a number of non-zero frequency components.

The above-mentioned image code transcoder measures the reduction rate from the amount of the first image code to the amount of the second image code, and the measured reduction rate is the target reduction rate. As described above, a control unit that changes the number of non-zero frequency components to be zero may be further included.

According to a fifth aspect of the present invention, a decoding means for decoding the first image code to obtain the frequency component of each coded block of the image, and each decoding block when the coded block is re-coded To reduce the amount of code per coding block,
Amplitude changing means for reducing or zeroing the amplitude of all or part of the frequency components of each coding block, and the amplitude of one or more frequency components of which the amplitude has become zero by the amplitude changing means. And non-zero means, and re-encoding means for re-encoding the frequency components of each block in which the amplitude of all or part of the frequency components has decreased or became zero to obtain a second image code. A code transcoder is provided.

The above-described image code transcoder measures the reduction rate from the amount of the first image code to the amount of the second image code, and the measured reduction rate is the target reduction rate. The control means for controlling the amplitude changing means may further be provided.

[0049]

DETAILED DESCRIPTION OF THE INVENTION Embodiments of the present invention will be described in detail below with reference to the drawings.

FIG. 1 shows the structure of a transcoder according to an embodiment of the present invention. FIG. 2 shows a format of an input / output signal of the transcoder according to the embodiment of the present invention. The transcoder shown in FIG. 1 receives the original program stream 151 in the format shown in FIG. 2, transcodes it, and outputs the transcoded program stream 161 in the format shown in FIG. The program stream is defined by the MPEG2 standard. As apparent from FIG. 2, the transcoded program stream 161 has a slice (slice 1) when compared with the original program stream 151.
6-6 are different in that they are shorter, and other parts are the same.

Referring to FIG. 1, a transcoder according to an embodiment of the present invention is a video PES (Packetized Elemen).
tary Stream) detection unit 101, video PES transcoder 102, transcoded video PESFIFO1
03, PES counter 104, original program stream FIFO 105, D type flip-flop 106, subtractor 107, comparator 108, latch 109, transcoded video PES counter 110, comparator 111,
Rise detection circuit 112, RS flip-flop 11
3, the multiplexer 114 is provided.

The video PES detector 101 mainly detects the video PES in the original program stream 151. The video PES detector 101 receives the original program stream 151 and inputs the original program stream 151.
1. Video PE in original program stream 151
The video PES enable signal 171, which becomes active at the portion where S is detected, the PES start code detection signal which becomes active when a PES start code detection signal is detected from the original program stream 151, and the end of the video PES in the original program stream 151. The video PES end detection signal 173 which becomes active when is detected is output.

The video PES transcoder 102 reduces the amount of image code for each DCT block included in each slice in the video PES. This reduction shortens each slice and each video PES. The video PES shortened in this way is the transcoded video PES152. The video PES transcoder 102 receives the original program stream 151, the video PES enable signal 171, the video PES end detection signal 173, and the reduction rate instruction signal 176, and receives the transcoded video PES 152, the video PES end detection signal 173, and the transformer. The coded video PES enable signal 174 is output. The transcoded video PES enable signal 174 indicates that the transcoded video PES is valid for the video PES transcoder 102.
This signal is active when the ES 152 is being output. A timing diagram of input / output signals of the video PES transcoder 102 is shown in FIG. The video PES enable signal 171 becomes active in the video PES (video PES1, video PES2, video PES3) portion. The video PES end detection signal 173 is the video PES.
(Video PES1, Video PES2, Video PES3)
Become active at the end of. The transcoded video PES152 is the video P in the original program stream.
Compared to ES, each slice (slice 1 ', ...
The difference is that the slice 6 ') is shorter.
Transcoded video PES enable signal 174
Is active in a portion of each slice of the transcoded video PES that includes a valid image code and a portion other than the slice (user data 1, user data 2, user data 3). A detailed description of the structure and operation of the video PES transcoder 102 will be given later.

Returning to FIG. 1, the transcoded video PESFIFO 103 has the transcoded video P
Write the ES enable signal 174 to the write enable (WE
N) as a valid transcoded video PES and video PES end detection signal 173.
Only these are temporarily stored, and these are output according to a multiplex control signal (described later) output by the RS flip-flop 113. Therefore, transcoded video PES
The FIFO 103 has a function of removing a slice end without a video code in the video PES and a function of variably delaying the transcoded video PES.

The PES counter 104 receives the PES start code detection signal 172 and outputs a PES count 175 indicating the counted value.

Original program stream FIFO 105
Inputs the inverted signal of the video PES enable signal 171 as a write enable (WEN) signal, and according to this inverted signal, the original program stream 151 and PE
The S count 175 is temporarily stored. Therefore, as shown in FIG. 4, the original program stream FIFO 105 includes the PES (user data PES1, audio PES1, PES) other than the video PES in the original program stream 151.
The PES count (n, n + 3, n + 4) corresponding to the PES other than the audio PES2) and the video PES is temporarily stored. The original program stream FIFO 105 is
The temporarily stored data is output according to a multiplex control signal (described later) output by the RS flip-flop 113. Therefore, the original program stream FIFO 105
Is the video PE of the original program stream 151.
It has a function of removing S and a function of variably delaying the original program stream 151 from which the video PES is removed.

D type flip-flop 106, subtractor 107, comparator 108, latch 109, transcoded video PES counter 110, comparator 111, rising edge detection circuit 112 and RS flip-flop 113.
Is the above-mentioned multiplex control signal (multiplexer 1
Controls 14 and transcoded video PE
It is provided to generate a read enable (REN) for the SFIFO 103 and a read enable (REN) for the original program stream FIFO 105. These timing diagrams are shown in FIG.

First, the original program stream FIFO1
Since the REN 05 is active, the original program stream FIFO 105 outputs the PES count having the value n and the user data PES1. When the user data PES1 ends, the value of the PES count changes from n to n + 3. Therefore, 3 is output from the subtractor 107, the output of the comparator 108 becomes HIGH, and the latch 1
3 is latched in 09, and transcoded video P
The ES counter 110 is loaded with “1” and the RS flip-flop 113 is reset.

Since the RS flip-flop 113 is reset, the reading of the user data 1 from the original program stream FIFO 105 is stopped, and the video PES1 'is read from the transcoded video PESFIFO103. In addition, the output of the multiplexer 114 switches from the user data PES1 to the video PES1 ′. At the end of the video PES1 ', the video P read from the transcoded video PESFIFO103
Since the ES end detection signal becomes HIGH, the count value of the transcoded video PES counter 110 increases from 1 to 2. Next, transcoded video P
The video PES2 ′ is read from the ESFIFO 103. At the end of the video PES2 ′, the video end detection signal read from the transcoded video PESFIFO becomes HIGH again, so that the count value of the transcoded video PES counter 110 increases from 2 to 3.

When the count value of the transcoded video PES counter 110 reaches 3, this value and the latch 10
Since the value 3 held by 9 becomes equal, the comparator 11
The output of 1 becomes active (HIGH). The rising edge detection circuit 112 detects a change in the output of the comparator 111 and generates a set signal for the RS flip-flop 113. Therefore, since the RS flip-flop 113 is set at this time, the transcoded video PESFI is set.
The reading of the video PES2 ′ from the FO 103 is stopped, and the audio P from the original program stream FIFO 105 is stopped.
ES1 is read. In addition, the output of the multiplexer 114 switches from the video PES2 ′ to the audio PES1.

When the audio data PES1 ends,
The value of the PES count changes from n + 3 to n + 4, so that the subtractor 107 outputs 1 but the comparator 108
Output remains low. Therefore, the latch 109 remains latching 3, the count of the transcoded video PES counter 110 remains “3”, and the RS flip-flop 113 remains set. Therefore, next, the original program stream FIFO1
The audio PES2 is read from 05.

Since the subsequent operation is the same as above,
The description is omitted.

With this operation, the multiplexer 1
14 outputs a transcoded program stream 161 having a format as shown in FIG. The bit rate can be reduced by making the read clocks of the transcoded video PESFIFO 103 and the original program stream FIFO 105 slower than these write clocks.

Next, the video PES transcoder 102.
Will be described in detail.

FIG. 6 shows a video PES transcoder 10
An example of the structure of 2 is shown. Referring to FIG. 6, video PE
The S transcoder 102 includes a slice detector 121, a variable length decoder 122, a DCT coefficient changing unit 123, a multiplexer 124, a variable length decoder 125, and a logical product gate 12.
6 and a code amount control unit 127.

The slice detector 121 has the original program stream 151 and the video PES enable signal 171.
To detect the video start code and the slice from the video PES, and output the original program stream 151, the video PES enable signal 171, the video start code detection signal 180, and the slice enable signal 189. The video start code detection signal 180 is a signal that becomes active in a portion of the original program stream 151 output by the slice detection unit 121 that has a video start code. What is a video start code?
It is a start code having a value of 00H to B8H such as a picture start code (00H) and a slice start code (01H to AFH). Slice enable signal 18
Reference numeral 9 is a signal that becomes active in the slice portion of the original program stream 151 output by the slice detection unit 121.

The variable length decoder 122 receives the original program stream 151, the video PES enable signal 171, the video start code detection signal 180, and the slice enable signal 189, decodes the variable length code in the video PES, and outputs each DCT block. The pre-change decoded data 181 including the DCT coefficient of is output.

The DCT coefficient changing unit 123 inputs the DCT coefficient of each DCT block in the variable length decoded data 181, and reduces or zeros the amplitude of all or some of the DCT coefficients of each DCT block. DCT coefficients that are output and whose amplitude is not reduced or made zero are output as they are. Further, the DCT coefficient changing unit 123 changes the degree of reducing or zeroing the DCT coefficient according to the value of the coefficient changing signal 188. DCT coefficient changing unit 12
The method of reducing or zeroing the DCT coefficient performed by 3 will be described later in detail.

The multiplexer 124 inputs the DCT coefficient of each DCT block from the DCT coefficient changing section 123,
Data other than the DCT coefficient is input from the variable length decoder 122, and these input data are multiplexed and output.

The variable length encoder 125 is the multiplexer 1
The data multiplexed in 24 is variable-length coded, and the variable-length code is converted into the transcoded video PES15.
Output as 2. This variable length coding also includes variable length coding of the DCT coefficients of each DCT block. The variable length encoder 125 also outputs the transcoded video P to be output.
It also outputs a transcoded video PES enable signal 174 indicating whether or not each byte of ES 152 is valid. The transcoded video PES enable signal 174 shown in FIG. 3 is continuously active from the start point to the middle of the slice, which means that the variable length encoder 125 continuously outputs the variable length code from the beginning of the slice within the slice. It is when generated. Since the DCT coefficient of each DCT block is reduced or set to zero by the DCT coefficient changing unit 123, the variable length code is not generated from the middle to the end of the slice, and accordingly, the transcoding is performed in this portion. The do-video PES enable signal 174 becomes inactive. The variable length encoder 125 may intermittently generate the variable length code, and in this case, the transcoded video PES enable signal 174 is also intermittently active.

The logical product gate 186 takes the logical product of the inversion signal of the image code enable signal 185 and the video PES enable signal 171, and outputs this logical product as the image code disable signal 186.

Referring to FIG. 7, the code amount control unit 127
Includes an image code enable time measuring unit 201, an image code disable time measuring unit 202, a reduction rate measuring unit 203, and a coefficient change signal adjusting unit 204. The image code enable time measuring unit 201 uses the image code enable signal 18
5 is input, and the cumulative value T1 of the time when this signal is active is measured. Image code disable time measurement unit 2
02 receives the image code disable signal 186,
The cumulative value T2 of the time when this signal is active is measured. The reduction rate measuring unit 203 calculates the measurement reduction rate R by the following equation, R
= T2 / (T1 + T2), and the measurement reduction rate signal 213 indicating this is output. Coefficient change signal adjustment unit 204
Inputs the reduction rate instruction signal 176 and the measurement reduction rate signal 213, and the coefficient change signal 188 is set so that the measurement reduction rate R indicated by the measurement reduction rate signal 213 becomes the target reduction rate indicated by the reduction rate instruction signal 176. Change the value of.

Next, DC performed by the DCT coefficient changing unit 123
A method of reducing or zeroing the T coefficient will be described in detail.

The DCT coefficient changing unit 123 receives each D
DC when the DCT coefficient of the CT block is re-encoded
Change so that the code amount per T block is reduced. Examples of the changing method include (1) to
The method of (7) is used. That is, the DCT coefficient changing unit 1
23 reduces the amplitude of all or part of the input non-zero DCT coefficients to a smaller amplitude by using the method of any one of (1) to (7), or sets the amplitude to 0, thereby each DCT. Reduce the amount of code per block.

(1) A low pass filter is applied to the DCT coefficients of each DCT block to reduce or eliminate the amplitude of the high frequency DCT coefficients.

For example, a two-dimensional low pass filter having a frequency characteristic as shown in FIG. 8 is prepared. The frequency characteristic of this two-dimensional low pass filter is divided into four frequency regions (region A, region B, region C and region D).
The gain in each frequency region is 1 (region A), 3/4 (region B), 2/4 (region C) or 1/4 (region D).
The two-dimensional low pass filter multiplies each DCT coefficient by this gain and rounds down for integerization. When this two-dimensional filter is applied to the DCT coefficient f (i, j) shown on the left of FIG. 9, for example, the DCT coefficient F (i, j) shown on the right of FIG. 9 is obtained. When this two-dimensional low-pass filter is applied, f (2,3) = 1 of the non-zero DCT coefficient f (i, j)
Changes to F (2,3) = 0 but other non-zero coefficients f
(I, j) does not become zero but the amplitude decreases.

The two-dimensionally distributed DCT coefficients are arranged into one-dimensional DCT coefficients by zigzag scanning as shown in FIG. That is, the DCT coefficient is f (0,0),
f (0,1), f (1,0), f (2,0), f (1,
1), f (0,2) ,. ． ,. One-dimensionally arranged in the order of f (7, 7). These DCT coefficients are coded as a pair with the amplitude (level) of the non-zero DCT coefficient and the consecutive number of zero DCT coefficients (zero run length) preceding the non-zero DCT coefficient when arranged in this way. Be converted. A variable length code having a length according to the occurrence frequency is assigned to each pair of level and zero run length. The relationship between the level of each pair, the combination of zero run lengths, and the length of the variable length code is as shown in FIGS. 11 and 12. FIG. 11 shows a non-intra block and an intra block with intra_vlc_format = 0, and FIG.
This is for an intra block with tra_vlc_format = 1. However, FIGS. 11 and 12 show the relationship between the combination of the level and the zero run length and the length of the variable length code only for a part of all the combinations of the level and the zero run length.

The block shown in FIG. 9 is intra_vlc_format = 1.
F (i, j) before applying such a coding to a two-dimensional low-pass filter.
For the two-dimensional matrix on the left side of FIG. 9, for example, the amplitude (level) of the DCT coefficient f (1,1)
Is 1, and the corresponding zero run length is 3 when counting zeros by the zigzag scan of FIG. 10, and the corresponding code length is the value at level 1 and zero run length 3 in the variable length code length table of FIG. There will be 6.

Zero run lengths and code lengths corresponding to the amplitudes of other DCT coefficients are also obtained as shown below the two-dimensional matrix on the left side of FIG. Therefore, EOB (End Of B
The code length per block excluding lock) is 123 bits.

On the other hand, the block shown in FIG. 9 is intra_vlc_fo.
If it is an intra block with rmat = 1, a DCT after such a coding is applied to a two-dimensional low-pass filter
When the coefficient F (i, j) is calculated, the code length is obtained as shown below the two-dimensional matrix on the right side of FIG.
The code length per block excluding B is 109 bits.

Therefore, in the case of the intra block of intra_vlc_format = 1, as shown in FIG. 9, each non-zero DCT
The code amount per DCT block when the coefficient and the preceding zero-length are two-dimensionally Huffman-encoded along the zigzag scan shown in FIG. 10 is 123 bits before the two-dimensional filter is applied, but the two-dimensional filter After multiplying, it is reduced to 109 bits.

(2) The amplitude of the DCT coefficient of each DCT block is divided by an integer greater than 1, the quotient is rounded down to an integer, and the result of the integer conversion is multiplied by the integer. That is, when the amplitude before truncation is Amp1 and the amplitude after truncation is Amp2, Amp2 = int (Amp1 / n) × n, where n is an integer greater than 1. The value of n may be changed for each frequency of the DCT coefficient. For example, the value of n may be increased as the frequency of the DCT coefficient is higher. Further, although the above method may be performed for all DCT coefficients, such processing may not be performed for some DCT coefficients.

For example, as shown in FIG. 13, the frequency domain is divided into four frequency domains (area A, area B, area C and area D). The DCT coefficient of region A is divided by 2, rounded down to an integer, and then multiplied by 2. Similarly, the DCT coefficient in region B is divided by 4, then rounded down to an integer, then multiplied by 4, and the DCT coefficient in region C is 8.
Divide by, truncate to integer, then multiply by 8, divide by DCT coefficient in region D by 16, truncate to integer, then multiply by 16. When this method is applied to the DCT coefficient f (i, j) shown on the left of FIG. 14, for example, the DCT coefficient F (i, j) shown on the right of FIG. 14 is obtained. Applying this method, the non-zero DCT coefficients f (i,
f (1,1), f (2,1), f (0,
4) f (2,3) and f (6,4) change to zero, but the other non-zero coefficients f (i, j) do not become zero but decrease in amplitude. In the case of an intra block of intra_vlc_format = 1, as shown in FIG. 14, each non-zero DCT coefficient and the preceding zero run length are two-dimensional Huffman-encoded along the zigzag scan shown in FIG. The code amount per bit is 123 bits before this method, but 10 bits after this method.
It is reduced to 1 bit.

(3) The amplitude of all or some of the non-zero DCT coefficients of each DCT block is varied with the non-zero DCT coefficient corresponding to the set of consecutive zeros preceding the non-zero DCT coefficient when zigzag scanned. The length of the long code is reduced or set to 0 so that the length is shortened by one step or several steps.

The length of the two-dimensional Huffman coding corresponding to a set of non-zero DCT coefficients and the preceding consecutive zeros is:
The non-intra block and the intra block with intra_vlc_format = 0 are as shown in FIG.
The intra block with vlc_format = 1 is as shown in FIG. However, FIG. 11 and FIG. 12 show the length of the two-dimensional Huffman code only for some (non-zero coefficient, the number of consecutive zeros preceding it) pairs.

Therefore, as can be seen from FIG. 12, for example, in the case of the intra block of intra_vlc_format = 1, if the amplitude of the DCT coefficient which is preceded by one zero and whose amplitude is 6 is reduced to 5, it corresponds. The length of the Huffman code can be reduced from 14 bits to 9 bits. That is, in this way, the length of the corresponding Huffman code can be shortened by one step. When such a method is applied to the DCT coefficient f (i, j) of the intra block shown on the left of FIG. 15, for example, the DCT coefficient F (i, j) shown on the right of FIG.
Is obtained. In the example of FIG. 15, referring to the table of FIG. 12, DC is set so that the length of the Huffman code is shortened by one step.
The amplitude of the T coefficient f (i, j) is reduced to the DCT coefficient F (i, j). In the case of the intra block of intra_vlc_format = 1, as shown in FIG. 15, each non-zero DCT coefficient and the preceding zero run length are two-dimensional Huffman-encoded DCT blocks along the zigzag scan shown in FIG. The code amount per hit is 123 bits before this method is performed, but is reduced to 90 bits after this method.

In the example of FIG. 15, the amplitude of each non-zero DCT coefficient is reduced so that the length of the Huffman code is shortened by one step, but each non-zero DCT coefficient is reduced by several steps. The amplitude of the zero DCT coefficient may be reduced or even zeroed.

(4) In (3), the decrease rate of the amplitude is set not to exceed the predetermined rate. That is, when the amplitude before change is Amp3 and the amplitude after change is Amp4, (Amp3-Amp4) / Amp3 <r, where r is a number greater than 0 and less than 1. The values of r are, for example, 0.9, 0.8, 0.7,
0.6, 0.5, etc. As described above, by preventing the decrease rate of the amplitude from exceeding the predetermined rate, it is possible to suppress the image quality deterioration.

FIG. 16 shows the case of r = 0.5 in the example shown in FIG.
3 shows a change in the amplitude of the DCT coefficient and a change in the amount of code per DCT block when the restriction is applied. intra_vlc_format
In the case of an intra block of = 1, as shown in FIG. 16, each non-zero DCT coefficient and the preceding zero run length are two-dimensional Huffman encoded along the zigzag scan shown in FIG. In the case of an intra block, the code amount of is 123 bits before performing this method, but after performing this method, it is 11 bits.
Reduced to 0 bits.

(5) In (3), the decrease value of the amplitude is set not to exceed the predetermined value. That is, when the amplitude before change is Amp3 and the amplitude after change is Amp4, (Amp3-Amp4) <diff, where diff is an integer exceeding 0, the value of diff is, for example, 1, 2, 3 , ... and so on. As described above, by preventing the decrease value of the amplitude from exceeding the predetermined value, it is possible to suppress the image quality deterioration.

FIG. 17 shows an example shown in FIG. 15 in which diff =
Amplitude change of DCT coefficient and DCT when limit of 1 is applied
The change in the code amount per block is shown. intra_vlc_form
In the case of the intra block of at = 1, as shown in FIG. 17, each non-zero DCT coefficient and the preceding zero run length are two-dimensional Huffman-encoded along the zigzag scan shown in FIG. In the case of an intra block, the code amount per bit is 123 bits before performing this method, but is 11 after performing this method.
It is reduced to 9 bits.

(6) The amplitude of the final predetermined number s of non-zero DCT coefficients when the DCT coefficients are zigzag scanned is set to zero.

For example, the value of the predetermined number s is set to 2, and FIG.
When this method is applied to the DCT coefficient shown on the left of FIG.
The DCT coefficient as shown on the right of is obtained. intra_vlc_fo
In the case of the intra block of rmat = 1, as shown in FIG. 18, each non-zero DCT coefficient and the zero length preceding it are two-dimensionally Huffman-encoded along the zigzag scan shown in FIG. The code amount per bit is 123 bits before this method,
After performing this method, it is reduced to 75 bits.

(7) The methods of (1) to (6) are combined.

For example, by applying the method (1) to the DCT coefficients shown on the left side of FIG. 14, the right side of FIG. 14 (or the left side of FIG. 19).
When the method of (4) is further applied after obtaining the DCT coefficient shown in (4), the DCT coefficient shown on the right of FIG. 19 is obtained. in
In case of intra block of tra_vlc_format = 1,
4 and FIG. 19, DC when each non-zero DCT coefficient and the zero length preceding it are two-dimensionally Huffman encoded along the zigzag scan shown in FIG.
The code amount per T block is 1 before performing this method.
It is 23 bits, but is reduced to 90 bits after performing this method.

(8) In the method of (1) to (3) or (7), in order to reduce the total code amount per DCT block, it was originally a non-zero DCT coefficient, but (1) to
All or part of the DCT coefficients which have become zero by the method (3) or (6) are made into non-zero DCT coefficients.

In the example of the method (1) shown in FIG. 9, f
Since (2,3) = 1 is changed to F (2,3) = 0, the code length 6 and f of the code corresponding to f (2,3) and the preceding zero run length (= 2) are From the sum (= 13) of the code lengths 7 of the codes corresponding to the non-zero DCT coefficient f (3,2) (= 7) after (2,3) and the preceding zero run length (= 0) Also, the code length (= 24) of the code of the corresponding area (the code corresponding to F (3,2) (= 5) and the corresponding zero run length (= 3))
Has become longer. The following measures are taken to solve this problem.

First, by the method of (1) to (3) or (7), the amplitude of the non-zero DCT coefficient is reduced or set to zero, the DCT block is coded, and the code amount per DCT block is obtained. Then, if there is a DCT coefficient that has changed from non-zero to zero by the method (1), that coefficient is made non-zero (for example, set to 1. It may be returned to the original amplitude), and the DCT block. Is encoded to obtain the code amount per DCT block. If the latter code amount is smaller, the DCT coefficient changed to zero is set to non-zero.

When the method (8) is applied to the method (1), the amplitude of F (2,3) is 1 as shown on the right side of FIG.
Becomes In this case, the code amount per DCT block is
This is 97 bits, which is smaller than the code amount of 109 bits in the example shown on the right side of FIG.

When the method (8) is applied to the method (2), the amplitude of F (2,3) is 1 as shown on the right side of FIG.
Becomes In this case, the code amount per DCT block is
It becomes 89 bits, and the code amount 101 in the example shown on the right side of FIG.
Less than a bit.

Although an example is not shown, it is also possible to apply the method (8) to the methods (3) and (6).

The DCT coefficient changing unit 123 adjusts the reduction rate of the amplitude of the DCT coefficient or the number of DCT coefficients to be changed according to the value of the coefficient changing signal 188, whereby the code amount of each DCT block is changed. Adjust the rate of reduction. When the method (1) is used, the rate of change in the amplitude of the DCT coefficient can be adjusted by changing the frequency characteristic of the low pass filter. When the method (2) is used, the rate of change in the amplitude of the DCT coefficient can be adjusted by changing the value of n for each frequency. When the method (3) is used, the number of DCT coefficients to be changed can be adjusted by changing the number of DCT coefficients to reduce the amplitude. Also, by changing the number of steps of reducing the length of the Huffman code, D
The ratio of the amplitude of the CT coefficient can be adjusted. (4)
When using the method of, the number of DC coefficients to be changed can be adjusted by changing the value of r.
When the method (5) is used, the number of DCT coefficients to be changed can be adjusted by changing the value of the predetermined number s. The DCT coefficient changing unit 123 appropriately changes the method to be used or a combination of these methods according to the value of the coefficient changing signal 188 to determine the reduction ratio of the amplitude of the DCT coefficient and / or the number of DCT coefficients to be changed. Can be adjusted.

Note that the configurations of the transcoder shown in FIG. 1, the video PES transcoder shown in FIG. 6, and the code amount control unit shown in FIG. 7 are merely examples realized by hardware. Further, in the transcoder shown in FIG. 1 and the video PES transcoder shown in FIG. 6, a circuit for delay adjustment in units of several clocks is omitted. The transcoder can be realized by a configuration other than that shown in FIG. In particular, the transcoder can also be realized by causing the computer to read and execute a program for causing the computer to function as the transcoder. Therefore, the computer can be caused to function as a transcoder by causing the computer to read and execute the program describing the method performed by the transcoder.

[0104]

As described above, according to the present invention,
Since the transcoding is performed by reducing or zeroing the amplitude of the DCT coefficient of each DCT block, the amount of calculation for transcoding can be reduced, and thus the computer can perform many transcodings at the same time. It will be possible.

Further, according to the present invention, the amplitude of the DCT coefficient does not change much without adopting a method of simply eliminating the high frequency component of the DCT coefficient of each DCT block. It is possible to obtain a decoded image corresponding to the bit rate from the generated image code.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of a transcoder according to an embodiment of the present invention.

FIG. 2 is a diagram showing a format of an input / output signal of the transcoder according to the embodiment of the present invention.

FIG. 3 is a timing diagram of input / output signals of a video PES transcoder of a transcoder according to an exemplary embodiment of the present invention.

FIG. 4 is a timing diagram of input signals of an original program stream FIFO of a transcoder according to an exemplary embodiment of the present invention.

FIG. 5 is a timing diagram of output signals and related signals of a transcoded video PESFIFO and an original program stream FIFO of a transcoder according to an exemplary embodiment of the present invention.

FIG. 6 is a block diagram showing a configuration of a video PES transcoder of the transcoder according to the embodiment of the present invention.

FIG. 7 is a block diagram showing a configuration of a code amount control unit of the transcoder according to the embodiment of the present invention.

FIG. 8 is a DC of a transcoder according to an embodiment of the present invention.
It is a figure for demonstrating the 1st method which a T coefficient change part performs.

FIG. 9 is a transcoder DC according to an embodiment of the present invention.
FIG. 6 is a diagram in which the T coefficient changing unit compares the DCT coefficient before performing the first method with the DCT coefficient and the DCT coefficient after performing the first method.

FIG. 10 is a diagram showing an example of zigzag scanning used when two-dimensional variable-length coding of DCT coefficients is performed.

FIG. 11 is a first chart showing the code length of a two-dimensional variable length code according to the MPEG standard.

FIG. 12 is a second chart showing the code length of a two-dimensional variable length code according to the MPEG standard.

FIG. 13 is a transcoder D according to an embodiment of the present invention.
It is a figure for demonstrating the 2nd method which a CT coefficient change part performs.

FIG. 14 is a transcoder D according to an embodiment of the present invention.
FIG. 6 is a diagram in which the CT coefficient changing unit compares the DCT coefficient before performing the second method with the DCT coefficient and after performing the second method.

FIG. 15 is a transcoder D according to an embodiment of the present invention.
FIG. 9 is a diagram in which the CT coefficient changing unit compares the DCT coefficient before performing the third method with the DCT coefficient and after performing the third method.

FIG. 16 is a transcoder D according to an embodiment of the present invention.
FIG. 10 is a diagram in which the CT coefficient changing unit compares the DCT coefficient before performing the fourth method and the DCT coefficient after performing the fourth method for the DCT coefficient.

FIG. 17 is a transcoder D according to an embodiment of the present invention.
It is a figure in which a CT coefficient change part compares a DCT coefficient before performing a 5th method with a DCT coefficient after performing it to a DCT coefficient.

FIG. 18 is a transcoder D according to an embodiment of the present invention.
It is a figure in which a CT coefficient change part compares a DCT coefficient before performing a 6th method with a DCT coefficient after performing it to a DCT coefficient.

FIG. 19 is a transcoder D according to an embodiment of the present invention.
It is a figure in which a CT coefficient change part compares a DCT coefficient before performing a 7th method and a DCT coefficient after performing with respect to a DCT coefficient.

FIG. 20 is a transcoder D according to an embodiment of the present invention.
It is a figure in which a CT coefficient change part compares a DCT coefficient before performing the 1st and 8th methods with a DCT coefficient after performing it to a DCT coefficient.

FIG. 21 is a transcoder D according to an embodiment of the present invention.
It is a figure which a CT coefficient change part compares the DCT coefficient before performing the 2nd and 7th method with respect to a DCT coefficient, and the DCT coefficient after performing it.

FIG. 22 is a block diagram showing the structure of a conventional MPEG decoder.

FIG. 23 is a block diagram showing a configuration of an MPEG encoder according to a conventional example.

FIG. 24 is a block diagram showing a configuration of a transcoder according to a conventional example.

FIG. 25 is a graph showing an example of characteristics of a quantization scale code and a quantization scale according to the MPEG2 standard.

[Fig. 26] Fig. 26 is a diagram for describing a conventional method of reducing the code amount by thinning out P pictures.

FIG. 27 is a diagram for explaining a conventional method for reducing the code amount by converting an I frame into a P frame.

FIG. 28 thins out I and P pictures, and
It is a figure for demonstrating the conventional method of reducing a code amount by converting an I frame into a P frame.

[Explanation of symbols]

101 Video PES detector 102 Video PES transcoder 103 transcoded video PESFIFO 104 PES counter 105 Original program stream FIFO 106 D type flip-flop 107 Subtractor 108 Comparator 109 latch 110 transcoded video PES counter 111 comparator 112 Rise detection circuit 113 RS flip-flop 114 multiplexer

   ─────────────────────────────────────────────────── ─── Continued front page    F term (reference) 5C059 KK15 KK41 MA00 MA23 MC11                       MC38 ME01 SS02 SS13 TA69                       TB08 TC38 TD06 TD07 UA02                       UA05 UA12                 5J064 AA01 BA09 BB05 BC01 BC08                       BC11 BC16 BC18 BC23 BD03

Claims (48)

[Claims] 1. Decoding means for decoding a first image code to obtain a frequency component of each coded block of an image, and a code amount for each coded block when each coded block is recoded. So as to reduce or zero the amplitude of all or part of the frequency components of each encoded block, and the amplitude changing means of each block where all or part of the frequency components have their amplitude reduced or become zero. An image code transcoder comprising: a re-encoding unit for re-encoding a frequency component to obtain a second image code, wherein the amplitude changing unit applies a low-pass filter to the frequency component of each block. Image code transcoder to do. 2. The image code transcoder according to claim 1, further comprising means for making non-zero the amplitude of one or more frequency components among the frequency components whose amplitude has become zero by said amplitude changing means. An image code transcoder, comprising: 3. The transcoder for the image code according to claim 1, further comprising a measuring unit that measures a reduction rate from the amount of the first image code to the amount of the second image code. And a control means for changing the frequency characteristic of the low-pass filter so that the reduction rate becomes a target reduction rate. 4. Decoding means for decoding a first image code to obtain a frequency component of each coded block of an image, and a code amount for each coded block when each coded block is recoded. So as to reduce or zero the amplitude of all or part of the frequency components of each encoded block, and the amplitude changing means of each block where all or part of the frequency components have their amplitude reduced or become zero. In a transcoder for an image code, comprising: a re-encoding means for re-encoding a frequency component to obtain a second image code, wherein the amplitude changing means divides all or some of the frequency components by an integer greater than 1. A transcoder for an image code, characterized in that the quotient is rounded down to an integer, and the result of the integer conversion is multiplied by the integer. 5. The image code transcoder according to claim 4, further comprising means for making non-zero the amplitude of one or more frequency components among the frequency components whose amplitude has become zero by the amplitude changing means. An image code transcoder, comprising: 6. The transcoder for the image code according to claim 4 or 5, further comprising: a measuring unit that measures a reduction rate from the amount of the first image code to the amount of the second image code. An image code transcoder, further comprising: a control unit that changes the integer so that the reduction rate becomes a target reduction rate. 7. Decoding means for decoding the first image code to obtain the frequency component of each coded block of the image, and reducing the code amount per coded block when each coded block is recoded. So as to reduce or zero the amplitude of all or part of the frequency components of each encoded block, and the amplitude changing means of each block where all or part of the frequency components have their amplitude reduced or become zero. Recoding means for recoding frequency components to obtain a second image code, and a transcoder for the image code comprising: the amplitude changing means, wherein the amplitude changing means scans the frequency components of each coded block Reducing or zeroing the amplitude of the non-zero frequency component such that the length of the variable length code corresponding to the number of components and zeros preceding the non-zero frequency component is reduced by one or several steps. Video code transcoder characterized. 8. The image code transcoder according to claim 7, wherein a non-zero frequency component that reduces or reduces amplitude is
The amplitude of the non-zero frequency component and the amplitude of the variable length code corresponding to the number of zeros preceding the non-zero frequency component after being reduced to be short compared to the amplitude before being reduced to such An image code transcoder characterized in that the reduction rate is limited to a predetermined rate or less. 9. The image code transcoder according to claim 7, wherein a non-zero frequency component that reduces or zeros the amplitude,
The amplitude after the variable length code corresponding to the non-zero frequency component and the number of zeros preceding the non-zero frequency component after being reduced to a shorter length was subtracted from the amplitude before being so reduced. An image code transcoder characterized in that the difference is limited to a predetermined value or less. 10. The image code transcoder according to claim 7, further comprising means for making the amplitude of one or more frequency components among the frequency components whose amplitude has become zero by the amplitude changing means non-zero. An image code transcoder, comprising: 11. The image code transcoder according to claim 7, wherein the measurement is performed by measuring a reduction rate from the amount of the first image code to the amount of the second image code. A transcoder for an image code, further comprising: a control unit configured to change the number of frequency components whose amplitude is reduced or made zero so that the measured reduction rate becomes a target reduction rate. . 12. The image code transcoder according to claim 7, wherein the reduction rate from the amount of the first image code to the amount of the second image code is measured. And a control means for changing the degree to which the amplitude is reduced or made zero so that the measured reduction rate becomes a target reduction rate. 13. Decoding means for decoding a first image code to obtain a frequency component of each coded block of an image, and a code amount for each coded block when each coded block is recoded. So as to reduce or zero the amplitude of all or part of the frequency components of each encoded block, and the amplitude changing means of each block where all or part of the frequency components have their amplitude reduced or become zero. Recoding means for recoding frequency components to obtain a second image code, and a transcoder for the image code, wherein the amplitude changing means is one from the end when the frequency components of each coding block are scanned. An image code transcoder characterized in that a number of non-zero frequency components equal to or more than zero is set to zero. 14. The image code transcoder according to claim 13, wherein the reduction rate from the amount of the first image code to the amount of the second image code is measured, and the measured reduction is performed. A transcoder for an image code, further comprising: control means for changing the number of non-zero frequency components to be zero so that the rate becomes a target reduction rate. 15. Decoding means for decoding the first image code to obtain the frequency component of each coded block of the image, and reducing the code amount per coded block when each coded block is recoded. As described above, one or more of an amplitude changing unit that reduces or zeros the amplitude of all or a part of the frequency components of each encoded block, and one or more of the frequency components whose amplitude has become zero by the amplitude changing unit. A means for making the amplitude of the frequency component non-zero, and a re-encoding means for re-encoding the frequency component of each block in which the amplitude of all or part of the frequency component has decreased or has become zero to obtain a second image code. , An image code transcoder comprising. 16. The image code transcoder according to claim 15, wherein a measuring unit that measures a reduction rate from the amount of the first image code to the amount of the second image code; A transcoder for an image code, further comprising: control means for controlling the amplitude changing means so that the reduction rate becomes a target reduction rate. 17. A decoding step of decoding a first image code to obtain a frequency component of each coded block of an image, and a code amount for each coded block is reduced when each coded block is recoded. So as to reduce the amplitude of all or some of the frequency components of each coded block to zero, and to change the amplitude of all or some of the frequency components of each block to zero. A re-encoding step of re-encoding the frequency component to obtain a second image code, and a trans-coding method of the image code, comprising: in the amplitude changing step, a low-pass filter is applied to the frequency component of each block. Image code transcoding method. 18. The method of transcoding an image code according to claim 17, further comprising the step of making the amplitude of one or more frequency components among the frequency components whose amplitude becomes zero in the amplitude changing step non-zero. A method for transcoding an image code, further comprising: 19. The transcoding method for an image code according to claim 17, further comprising a measuring step of measuring a reduction rate from the amount of the first image code to the amount of the second image code. And a control step of changing the frequency characteristic of the low-pass filter so that the reduction rate thus obtained becomes a target reduction rate. 20. A decoding step of decoding a first image code to obtain a frequency component of each coded block of an image, and a code amount for each coded block is reduced when each coded block is recoded. So as to reduce the amplitude of all or some of the frequency components of each coded block to zero, and to change the amplitude of all or some of the frequency components of each block to zero. A recoding step of recoding frequency components to obtain a second image code, and a transcoding method of the image code having: in the amplitude changing step, all or some of the frequency components are divided by an integer greater than 1. Then, the quotient is rounded down to an integer, and the result of the integer conversion is multiplied by the integer. 21. The image code transcoding method according to claim 20, further comprising the step of making the amplitudes of one or more frequency components among the frequency components whose amplitudes become zero in the amplitude changing step non-zero. A method for transcoding an image code, further comprising: 22. The image code transcoding method according to claim 20 or 21, further comprising a measuring step of measuring a reduction rate from the amount of the first image code to the amount of the second image code. And a control step of changing the integer so that the obtained reduction rate becomes a target reduction rate. 23. A decoding step of decoding a first image code to obtain a frequency component of each coded block of an image, and a code amount for each coded block is reduced when each coded block is recoded. So as to reduce the amplitude of all or some of the frequency components of each coded block to zero, and to change the amplitude of all or some of the frequency components of each block to zero. A recoding step of recoding frequency components to obtain a second image code, and a transcoding method of the image code having: wherein in the amplitude changing step, a non-zero value is obtained when the frequency components of each coded block are scanned. The non-zero frequency component is oscillated so that the length of the variable-length code corresponding to the number of zeros preceding the frequency component and the non-zero frequency component is shortened by one step or several steps. Image transcoding method code, characterized by the reduced thereby or zero. 24. The transcoding method of the image code according to claim 23, wherein a non-zero frequency component that reduces or zeros the amplitude,
The amplitude of the non-zero frequency component and the amplitude of the variable length code corresponding to the number of zeros preceding the non-zero frequency component after being reduced to be short compared to the amplitude before being reduced to such A method for transcoding an image code, wherein the reduction rate is limited to a predetermined rate or less. 25. The transcoding method for an image code according to claim 23, wherein the non-zero frequency component that reduces or reduces the amplitude to:
The amplitude after the variable length code corresponding to the non-zero frequency component and the number of zeros preceding the non-zero frequency component after being reduced to a shorter length was subtracted from the amplitude before being so reduced. A method of transcoding an image code, wherein the difference is limited to a value less than a predetermined value. 26. The image code transcoding method according to claim 23, further comprising the step of making the amplitude of one or more frequency components among the frequency components whose amplitudes become zero in the amplitude changing step non-zero. A method for transcoding an image code, further comprising: 27. The image code transcoding method according to claim 23, wherein a reduction rate from the amount of the first image code to the amount of the second image code is measured. A transformer for an image code, further comprising: a measuring step; and a control step of changing the number of frequency components whose amplitude is reduced or becomes zero so that the measured reduction rate becomes a target reduction rate. Coding method. 28. The image code transcoding method according to claim 23, wherein a reduction rate from the amount of the first image code to the amount of the second image code is measured. A method of transcoding an image code, further comprising: a step of controlling a degree of decreasing an amplitude or changing the amplitude to zero so that the measured reduction rate becomes a target reduction rate. 29. A decoding step of decoding a first image code to obtain a frequency component of each coded block of an image, and a code amount for each coded block is reduced when each coded block is recoded. So as to reduce the amplitude of all or some of the frequency components of each coded block to zero, and to change the amplitude of all or some of the frequency components of each block to zero. A recoding step of recoding frequency components to obtain a second image code, and a transcoding method of the image code having the: A method for transcoding an image code, wherein zero or more non-zero frequency components are set to zero. 30. The image code transcoding method according to claim 29, wherein a step of measuring a reduction rate from the amount of the first image code to the amount of the second image code is measured. A method of transcoding an image code, further comprising: a control step of changing the number of non-zero frequency components to be zero so that the reduction rate becomes a target reduction rate. 31. A decoding step of decoding a first image code to obtain a frequency component of each coded block of an image, and a code amount for each coded block is reduced when each coded block is recoded. So as to reduce the amplitude of all or part of the frequency components of each coding block or to zero, and one or more of the frequency components whose amplitude has become zero in the amplitude changing step. A step of making the amplitude of the frequency component non-zero, and a step of re-encoding the frequency component of each block in which the amplitude of all or part of the frequency component is reduced or becomes zero to obtain a second image code. A method for transcoding an image code, comprising: 32. The transcoding method of the image code according to claim 31, wherein a measuring step of measuring a reduction rate from the amount of the first image code to the amount of the second image code is measured. And a control step of controlling the amplitude changing step so that the reduction rate becomes a target reduction rate. 33. A decoding step of decoding a first image code to obtain a frequency component of each coded block of an image, and a code amount for each coded block is reduced when each coded block is recoded. So as to reduce the amplitude of all or some of the frequency components of each coded block to zero, and to change the amplitude of all or some of the frequency components of each block to zero. A re-encoding step of re-encoding a frequency component to obtain a second image code, and a trans-coding method of the image code, comprising: applying a low-pass filter to the frequency component of each block in the amplitude changing step. A program for causing a computer to execute a transcoding method of an image code characterized by. 34. The program of the image code according to claim 33, wherein the transcoding method of the image code is an amplitude of one or more frequency components among frequency components whose amplitude becomes zero in the amplitude changing step. A program further comprising the step of: 35. The program according to claim 33, wherein the transcoding method for the image code measures a reduction rate from the amount of the first image code to the amount of the second image code. A program, further comprising: a control step of changing a frequency characteristic of the low-pass filter so that the measured reduction rate becomes a target reduction rate. 36. A decoding step of decoding a first image code to obtain a frequency component of each coded block of an image, and a code amount for each coded block is reduced when each coded block is recoded. So as to reduce the amplitude of all or some of the frequency components of each coded block to zero, and to change the amplitude of all or some of the frequency components of each block to zero. A re-encoding step of re-encoding a frequency component to obtain a second image code, and a trans-coding method of an image code, comprising: in the amplitude changing step, all or some of the frequency components , The quotient is rounded down to an integer, and the result of the integerization is multiplied by the integer to cause a computer to execute a transcoding method of the image code. Program for. 37. The program according to claim 36, wherein the transcoding method of the image code sets the amplitude of one or more frequency components among the frequency components whose amplitude is zero in the amplitude changing step to non-zero. A program further comprising the step of: 38. The program according to claim 36, wherein the transcoding method for the image code measures a reduction rate from the amount of the first image code to the amount of the second image code. A program further comprising: a control step of changing the integer so that the measured reduction rate becomes a target reduction rate. 39. A decoding step of decoding a first image code to obtain a frequency component of each coded block of an image, and a code amount for each coded block is reduced when each coded block is recoded. So as to reduce the amplitude of all or some of the frequency components of each coded block to zero, and to change the amplitude of all or some of the frequency components of each block to zero. A recoding step of re-encoding frequency components to obtain a second image code, and a transcoding method of the image code, comprising: in the amplitude changing step, when the frequency components of each coded block are scanned, The non-zero frequency component is oscillated so that the length of the variable-length code corresponding to the number of zero frequency components and zeros preceding the non-zero frequency component is shortened by one step or several steps. Program for executing the transcoding method of image coding characterized a computer to a reduced or zero. 40. The program according to claim 39, wherein a non-zero frequency component that reduces or reduces the amplitude is
The amplitude of the non-zero frequency component and the amplitude of the variable length code corresponding to the number of zeros preceding the non-zero frequency component after being reduced to be short compared to the amplitude before being reduced to such A program characterized in that the reduction rate is limited to a predetermined rate or less. 41. The program of the image code according to claim 39, wherein a non-zero frequency component that reduces or reduces amplitude is
The amplitude after the variable length code corresponding to the non-zero frequency component and the number of zeros preceding the non-zero frequency component after being reduced to a shorter length was subtracted from the amplitude before being so reduced. A program characterized in that the difference is limited to a predetermined value or less. 42. The program according to claim 39, wherein the transcoding method for the image code sets the amplitude of one or more frequency components among the frequency components whose amplitude has become zero in the amplitude changing step to non-zero. A program further comprising the step of: 43. The program according to claim 39, wherein the transcoding method for the image code reduces the amount of the first image code to the amount of the second image code. A measurement step of measuring a rate, and a control step of changing the number of frequency components for decreasing or zeroing the amplitude so that the measured reduction rate becomes a target reduction rate. program. 44. The program according to any one of claims 39 to 42, wherein the transcoding method for the image code reduces the amount of the first image code to the amount of the second image code. A program, further comprising: a step of measuring a rate, and a control step of changing a degree of decreasing or zeroing an amplitude so that the measured reduction rate becomes a target reduction rate. 45. A decoding step of decoding a first image code to obtain a frequency component of each coded block of an image, and a code amount for each coded block is reduced when each coded block is recoded. So as to reduce the amplitude of all or some of the frequency components of each coded block to zero, and to change the amplitude of all or some of the frequency components of each block to zero. A re-encoding step of re-encoding a frequency component to obtain a second image code, and a trans-coding method of an image code, comprising the step of: To cause a computer to perform a method of transcoding an image code, characterized in that one or more non-zero frequency components from Program. 46. The program according to claim 45, wherein the method of transcoding the image code measures a reduction rate from the amount of the first image code to the amount of the second image code, A control step of changing the number of non-zero frequency components to be zero so that the measured reduction rate becomes a target reduction rate. 47. A decoding step of decoding a first image code to obtain a frequency component of each coded block of an image, and a code amount for each coded block is reduced when each coded block is recoded. So as to reduce the amplitude of all or part of the frequency components of each coding block or to zero, and one or more of the frequency components whose amplitude has become zero in the amplitude changing step. A step of making the amplitude of the frequency component non-zero, and a step of re-encoding the frequency component of each block in which the amplitude of all or part of the frequency component is reduced or becomes zero to obtain a second image code. And a program for causing a computer to execute a transcoding method of an image code having :. 48. The program according to claim 47, wherein the transcoding method of the image code measures a reduction rate from the amount of the first image code to the amount of the second image code. A program of an image code, further comprising: a control step of controlling the amplitude changing step so that the measured reduction rate becomes a target reduction rate.