KR950008958B1 - Quantization Apparatus for Discrete Cosine Transform in Dynamic Compensation Prediction Difference Signal - Google Patents
Quantization Apparatus for Discrete Cosine Transform in Dynamic Compensation Prediction Difference Signal Download PDFInfo
- Publication number
- KR950008958B1 KR950008958B1 KR1019920012973A KR920012973A KR950008958B1 KR 950008958 B1 KR950008958 B1 KR 950008958B1 KR 1019920012973 A KR1019920012973 A KR 1019920012973A KR 920012973 A KR920012973 A KR 920012973A KR 950008958 B1 KR950008958 B1 KR 950008958B1
- Authority
- KR
- South Korea
- Prior art keywords
- dct
- weighted
- quantizer
- output
- input
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
Classifications
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03M—CODING; DECODING; CODE CONVERSION IN GENERAL
- H03M7/00—Conversion of a code where information is represented by a given sequence or number of digits to a code where the same, similar or subset of information is represented by a different sequence or number of digits
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Compression Or Coding Systems Of Tv Signals (AREA)
- Compression, Expansion, Code Conversion, And Decoders (AREA)
Abstract
내용 없음.No content.
Description
제1도는 종래의 부호화 장치의 블럭 구성도.1 is a block diagram of a conventional encoding apparatus.
제2도는 본 발명에 따른 양자화 장치의 블럭 구성도.2 is a block diagram of a quantization device according to the present invention.
* 도면의 주요부분에 대한 부호의 설명* Explanation of symbols for main parts of the drawings
1,14 : 불연속 코사인 변환기(DCT) lA : 감산기1,14: Discrete Cosine Converter (DCT) lA: Subtractor
lB : 가산기 3 : 예비프레임lB: Adder 3: Preliminary frame
4 : 동적 에스티메이터(Mtion Estimator)4: Motion Estimator
5 : 프레임 메모리 7 : 역 불연속 코사인 변환기5: frame memory 7: inverse discontinuous cosine converter
8 : 가변길이 부호기(VLC) 9 : 버퍼8: variable length encoder (VLC) 9: buffer
11,13 : 웨이티드(Weighted) 양자화기 12,15 : 역웨이티드양자화기11,13: weighted quantizer 12,15: reverse weighted quantizer
본 발명은 동적 보상 예측 차신호(Motion Compensated Prediction Error Signal)에서의 불연속 코사인 변환기(DCT)에 대한 양자화 장치에 관한 것으로서 특히 웨이티드 양자화기를 사용하여 영상간 부호화시에 보다 높은 데이타 압축율과 재생화질 개선을 도모코자 한 것이다.The present invention relates to a quantization apparatus for a discrete cosine transform (DCT) in a motion compensated prediction error signal. In particular, a weighted quantizer improves higher data compression ratio and reproduction quality when inter-image encoding is performed. It was intended to promote.
종래의 부호화 장치는 제1도에 도시한 바와 같이 영상간 차신호의 양자화를 위해 모든 DCT 변환계수에 동일한 가중치를 설정한 후 균등양자화기(Unifonn Quantizer)(2)를 통해 균등양자화하게 된다.As shown in FIG. 1, the conventional encoding apparatus sets equal weights to all DCT transform coefficients for quantization of the difference signals between images, and then equalizes them through a uniform quantizer 2.
한편 영상내의 신호(Intraframe Signal)의 DCT 변환계수에서 낮은 주과수를 갖는 계수에 대하여서는 양자화 레벨수를 감소시킨다.On the other hand, the coefficient of the quantization level is reduced for the coefficient having a low main fruit tree in the DCT transform coefficient of the intraframe signal.
이것은 인간의 육안으로서 감지할 수 있는 저주파수를 갖는 계수에 대하여 보다 더 민감하게 적응하기 위한 것이다.This is to adapt more sensitively to coefficients with low frequencies that can be perceived by the human eye.
또한 DCT 변환계수를 지그재그형태로 스캔하여 가별길이 부호기(8)를 통해 부호화할 경우에 런(RUN)의 갯수를 증가시킬 수가 있기 때문에 보다 높은 압축율을 가지도록 한다.In addition, when the DCT transform coefficient is scanned in a zigzag form and encoded through the split length encoder 8, the number of runs can be increased, so that the compression ratio is higher.
이와같이 종래의 장치는 영상내 부호화시에는 웨이티드 양자화기/역양자화기를 사용하고, 영상간 부호화시에는 균등양자화기/역양자화기를 사용하고 있다.As described above, the conventional apparatus uses a weighted quantizer / dequantizer for intra-image encoding and an equal quantizer / dequantizer for inter-image encoding.
상기한 웨이티드 양자화는 MPEG(Moving Picture Expert Group)에서 제안한 바와 같이The weighted quantization described above is proposed by the Moving Picture Expert Group (MPEG).
F(k.l) :DCT 변환계수 k,1:0,l,···7F (k.l): DCT conversion factor k, 1: 0, l,
QF(k,l):양자화된 변환계수 W(k,1) : 가중치일 때QF (k, l): quantized transform coefficient W (k, 1): when weighted
8 16 19 22 26 27 29 348 16 19 22 26 27 29 34
16 16 22 24 27 29 34 3716 16 22 24 27 29 34 37
19 22 26 27 29 34 34 3819 22 26 27 29 34 34 38
22 22 26 27 29 34 34 3822 22 26 27 29 34 34 38
22 22 26 27 29 34 37 4022 22 26 27 29 34 37 40
22 26 27 29 32 35 40 4822 26 27 29 32 35 40 48
26 27 29 32 35 40 48 5826 27 29 32 35 40 48 58
26 27 29 34 38 46 56 6926 27 29 34 38 46 56 69
27 29 35 38 46 56 69 8327 29 35 38 46 56 69 83
이고 여기서 QF(k,l)=[16×F(k,l)]/W(k,l)/▲이다.Where QF (k, l) = [16 × F (k, l)] / W (k, l) / ▲.
상기한 ▲:버퍼상태에 따라 결정되는 상수이다.The above ▲: is a constant determined according to the buffer state.
즉, 종래 기술에서는 예측오차신호의 DCT 변환계수를 양자화할 때 모든 계수에 대하여 양자화 스텝크기를 균일하게 사용하므로 시각에 덜 민감한 성분도 양자화 과정에서 부호화 된다.That is, in the prior art, since the quantization step size is uniformly used for all coefficients when quantizing the DCT transform coefficients of the prediction error signal, less visually sensitive components are encoded during the quantization process.
그리고 DC 성분에 대해 이웃한 MB이 서로 다른 양자화 스템크기를 사용하면 저전송속도일수록 블럭효과가 생긴다.If the neighboring MBs use different quantization stem sizes for the DC component, a lower transmission rate causes a block effect.
그러나, 이것은 종래 기술은 영상내 부호화시에는 웨이티드 양자화기/역양자화기를 사용하고, 영상간 부호화시에는 균등양자화기/역양자화기를 사용하는 형태로 영상내 부호화시와 영상간 부호화시에 서로 다른 특성의 양자화기를 사용하기 때문에 영상화질의 저하를 초래하는 문제점이 있었다.However, this is because the prior art uses a weighted quantizer / inverse quantizer for intra-image encoding and an equal quantizer / dequantizer for inter-image encoding. Since the quantizer is used, there is a problem of causing degradation of the image quality.
본 발명은 이러한 종래 기술의 문제점을 해소고자 한 것으로서, 영상내 부호화시나 영상간 부호화시에 모두 동일한 특성을 갖는 웨이티드 양자화기를 사용하여 시각에 민감한 부분에 대해서는 세밀히 양자화하고, 둔감한 부분에 대해서는 덜 세밀히 양자화함으로써 영상간 부호화시에 같은 데이타량에 대해서 기존 장치보다 재생화질이 우수하도록 하고, 또한 DC 성분을 동일한 스텝크기로 양자화 함으로써 블럭효과가 줄어들도록 함을 그 목적으로 하는 것이다.The present invention solves the problems of the prior art, using a weighted quantizer having the same characteristics for both intra-image encoding and inter-image encoding, to finely quantize the visually sensitive portion and less for the insensitive portion. The purpose of the present invention is to provide finer quantization so that the reproduction quality is better than that of the existing apparatus for the same amount of data in inter-image encoding, and the block effect is reduced by quantizing the DC component to the same step size.
이하, 본 발명을 첨부된 도면에 따라 상세히 설명한다.Hereinafter, the present invention will be described in detail with reference to the accompanying drawings.
본 발명은 제2도에 나타낸 바와 같이 영상임력신호(IN)를 DCT 변환하는 DCT(1)를 거쳐 감산기(1A)의 일측 입력단에 연결하고, 상기 감산기(1A)의 출력단에는 웨이티드 양자화기(11)가 연결되며, 상기 웨이티드 양자화기(11)의 출력은 VLC(8)를 통해 버퍼(9)에 저장된 후 채널로 전송되는 동시에 역웨이티드 양자화기(15)에 입력되도록 연결하고, 상기 역웨이티드 양자화기(l5)를 거쳐 역웨이티드 양자화된 출력은 역웨이티드 양자화기(12)의 출력과 함께 가산기(1B)를 통해 더해진 후 역 DCT(7)에 입력되도록 연결하며, 상기 역 DCT(7)를 통해 역DCT 변환되어 재생된 영상출력은 프레임 메모리(5)를 거쳐 동적 에스티메이터(4)에서 영상입력신호(lN)와 비교되도록 연결하고, 상기 동적 에스티에이터(4)의 출력은 예비프레임(3)에 저장된 후 영상간 부호화시에 DCT(14)에 입력되고, 상기 DCT(14)의 출력은 웨이티드 양자화기(l34) 및 역웨이티드 양자화기(l2)를 순차적으로 거쳐 상기 감산기(1A)의 타측 입력단에 가해지도록 연결하여 구성된 것이다.As shown in FIG. 2, the present invention is connected to an input terminal of one side of the subtractor 1A via a DCT 1 for DCT conversion of the image force signal IN, and a weighted quantizer (A) at the output terminal of the subtractor 1A. 11) is connected, and the output of the weighted quantizer 11 is stored in the buffer 9 through the VLC (8) and then transmitted to the channel and connected to be input to the de-weighted quantizer 15, and The inverse weighted quantized output via the inverse weighted quantizer l5 is added through the adder 1B together with the output of the inverse weighted quantizer 12 and connected to be input to the inverse DCT 7. The video output reproduced by inverse DCT conversion through the DCT 7 is connected to the image input signal lN in the dynamic estimator 4 through the frame memory 5, and outputted from the dynamic estimator 4. Is stored in the preliminary frame 3 and input to the DCT 14 during inter-image encoding. The output of the DCT 14 is connected to be applied to the other input terminal of the subtractor 1A through the weighted quantizer 134 and the inverted weighted quantizer 1 2.
또한 본 발명은 상기 DCT(1)와 DCT(14)에 의하여 불연속 코사인 변환을 2부분으로 나누어 행하도록 구성된다.In addition, the present invention is configured to divide the discrete cosine transform into two parts by the DCT 1 and the DCT 14.
이와 같이 구성된 본 발명의 동작 및 작용효과를 설명하면 마음과 같다.Referring to the operation and effect of the present invention configured as described above is like mind.
주지하는 바와 같이 영상내의 부호화된 재생영상의 화질이 영상간의 부호화된 재생영상의 화질보다 양질이다.As is well known, the image quality of the encoded reproduced image in the image is higher than the image quality of the encoded reproduced image between the images.
본 발명은 이러한 영상간의 부호화된 재생영상이 영상내의 부호화된 재생영상과 동일한 화질을 유지토록함에 있다.The present invention is such that the encoded playback video between the images maintains the same image quality as the encoded playback video in the video.
첫번째, 영상내의 부호화시의 동작을 설명하면, DCT(1)를 거쳐 불연속 고사인 변환된 원래 영상을 F(k,l)이라 가정하면, F(k,l)=QF(k,l) △(k,1)이고 △(k,I)=△×W(k,1)/16이다.·First, the operation at the time of encoding in the image will be explained. F (k, l) = QF (k, l) Δ Assuming that the original image obtained by discontinuous sine conversion through DCT (1) is F (k, l). (k, 1) and Δ (k, I) = Δ × W (k, 1) / 16.
여기서 QF(k,1):정수, △(k,1) = 양자화 스텝크기를 나타낸다. 양자화 잡음 Eintra(k,1)는Where QF (k, 1): integer, Δ (k, 1) = quantization step size. Quantization Noise Eintra (k, 1) is
로 나타나고, 여기서 CF(k, l):실수이다.Where CF (k, l): is a real number.
두번째, 영상간의 부호화시의 동작을 설명하면, 본 발명에 의해 예측영상은 양자화되어 다음과 같은 변형된 예측신호를 갖는다.Second, the operation of encoding between images will be described. According to the present invention, a predicted image is quantized to have a modified predicted signal as follows.
여기서 QP(k,1):정수이다.Where QP (k, 1) is an integer.
원래 영상은 다음과 같이 표현된다.The original image is expressed as follows.
따라서, 변형된 예측오차는Therefore, the modified prediction error
여기서, 변형된 예측오차를 양자화하면Here, if the quantized modified prediction error
여기서 NINT{ } : Nearest Integer 값을 취한다.Where NINT {}: takes the Nearest Integer value.
따라서, 영상간의 부호화시에 재생된 영상은 다음과 같다.Therefore, the reproduced video at the time of encoding between the images is as follows.
즉, 영상간의 부호화시에 재생된 영상과 영상내의 부호화시에 재생된 영상의 화질이 동일하게 나타남을 알 수가 있다.In other words, it can be seen that the picture quality reproduced at the time of encoding between pictures and the picture reproduced at the time of encoding in the picture are the same.
이상에서와 같이 본 발명은 영상간 부호화시에 현재 영상에서 예측영상을 뺀 예측오차 DCT 변환하고 양자화함에 있어, 보다 데이타 압축율이 높은 웨이티드 양자화기를 사용함으로써 기존 장치에 비하여 재생화질이 우수할 뿐만 아니라 블럭효과가 줄어들게 할 수 있는 이점이 있는 것이다.As described above, the present invention is not only superior to the conventional apparatus by using a weighted quantizer having a higher data compression ratio in transforming and quantizing the prediction error DCT by subtracting the predictive image from the current image during inter-image encoding. There is an advantage that can reduce the block effect.
Claims (1)
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
KR1019920012973A KR950008958B1 (en) | 1992-07-21 | 1992-07-21 | Quantization Apparatus for Discrete Cosine Transform in Dynamic Compensation Prediction Difference Signal |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
KR1019920012973A KR950008958B1 (en) | 1992-07-21 | 1992-07-21 | Quantization Apparatus for Discrete Cosine Transform in Dynamic Compensation Prediction Difference Signal |
Publications (2)
Publication Number | Publication Date |
---|---|
KR940003196A KR940003196A (en) | 1994-02-21 |
KR950008958B1 true KR950008958B1 (en) | 1995-08-09 |
Family
ID=19336691
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
KR1019920012973A Expired - Fee Related KR950008958B1 (en) | 1992-07-21 | 1992-07-21 | Quantization Apparatus for Discrete Cosine Transform in Dynamic Compensation Prediction Difference Signal |
Country Status (1)
Country | Link |
---|---|
KR (1) | KR950008958B1 (en) |
-
1992
- 1992-07-21 KR KR1019920012973A patent/KR950008958B1/en not_active Expired - Fee Related
Also Published As
Publication number | Publication date |
---|---|
KR940003196A (en) | 1994-02-21 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US4723161A (en) | Method and arrangement of coding digital image signals utilizing interframe correlation | |
KR960006762B1 (en) | Efficient 2D Scan Selection Circuit for Image Coding | |
JP3814321B2 (en) | Quantization parameter determination apparatus and method | |
KR950002472A (en) | DC Transform Quantizer Using Human Visual Characteristics | |
JPH03204290A (en) | Video signal coding method | |
EP0519962A1 (en) | Digital image coding using a random scanning of image frames. | |
CA2109138A1 (en) | Device for encoding digital signals representing images, and corresponding decoding device | |
KR940020832A (en) | Adaptive Quantization Method of High-definition Television and System Coder Using the Same | |
JP3739388B2 (en) | Double reference image encoding method, encoding apparatus, and decoder | |
EP0680217A2 (en) | Video signal decoding apparatus capable of reducing blocking effects | |
US5946421A (en) | Method and apparatus for compensating quantization errors of a decoded video image by using an adaptive filter | |
KR20020001760A (en) | Image data compression | |
JP2003526237A (en) | Quantization method for bit rate transcoding applications | |
JPH0468988A (en) | High efficient coder for moving picture signal | |
JP4081727B2 (en) | Image encoding apparatus, image encoding method, recording apparatus, and recording method | |
EP0680218B1 (en) | Image signal decoding apparatus having an encoding error compensation | |
KR950008958B1 (en) | Quantization Apparatus for Discrete Cosine Transform in Dynamic Compensation Prediction Difference Signal | |
KR970004924B1 (en) | Improved motion vector transmission apparatus and method using layered coding | |
KR20030000310A (en) | Apparatus for transcoding video | |
JP2512165B2 (en) | Video signal encoder | |
KR100338725B1 (en) | Pictures coding method approximating coding bits to target coding bits and apparatus therefor | |
KR100213283B1 (en) | Moving picture encoder | |
KR0126482Y1 (en) | Data Loss Prevention Device of Image Compressor | |
KR20060034294A (en) | Encoding Method and Device | |
KR0148136B1 (en) | Inter/intramode code method and apparatus |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
A201 | Request for examination | ||
PA0109 | Patent application |
St.27 status event code: A-0-1-A10-A12-nap-PA0109 |
|
PA0201 | Request for examination |
St.27 status event code: A-1-2-D10-D11-exm-PA0201 |
|
R17-X000 | Change to representative recorded |
St.27 status event code: A-3-3-R10-R17-oth-X000 |
|
PG1501 | Laying open of application |
St.27 status event code: A-1-1-Q10-Q12-nap-PG1501 |
|
E902 | Notification of reason for refusal | ||
PE0902 | Notice of grounds for rejection |
St.27 status event code: A-1-2-D10-D21-exm-PE0902 |
|
AMND | Amendment | ||
P11-X000 | Amendment of application requested |
St.27 status event code: A-2-2-P10-P11-nap-X000 |
|
P13-X000 | Application amended |
St.27 status event code: A-2-2-P10-P13-nap-X000 |
|
E601 | Decision to refuse application | ||
PE0601 | Decision on rejection of patent |
St.27 status event code: N-2-6-B10-B15-exm-PE0601 |
|
J2X1 | Appeal (before the patent court) |
Free format text: APPEAL AGAINST DECISION TO DECLINE REFUSAL |
|
PB0901 | Examination by re-examination before a trial |
St.27 status event code: A-6-3-E10-E12-rex-PB0901 |
|
G160 | Decision to publish patent application | ||
PG1605 | Publication of application before grant of patent |
St.27 status event code: A-2-2-Q10-Q13-nap-PG1605 |
|
B701 | Decision to grant | ||
PB0701 | Decision of registration after re-examination before a trial |
St.27 status event code: A-3-4-F10-F13-rex-PB0701 |
|
GRNT | Written decision to grant | ||
PR0701 | Registration of establishment |
St.27 status event code: A-2-4-F10-F11-exm-PR0701 |
|
PR1002 | Payment of registration fee |
St.27 status event code: A-2-2-U10-U11-oth-PR1002 Fee payment year number: 1 |
|
PR1001 | Payment of annual fee |
St.27 status event code: A-4-4-U10-U11-oth-PR1001 Fee payment year number: 4 |
|
R18-X000 | Changes to party contact information recorded |
St.27 status event code: A-5-5-R10-R18-oth-X000 |
|
PR1001 | Payment of annual fee |
St.27 status event code: A-4-4-U10-U11-oth-PR1001 Fee payment year number: 6 |
|
PR1001 | Payment of annual fee |
St.27 status event code: A-4-4-U10-U11-oth-PR1001 Fee payment year number: 8 |
|
PR1001 | Payment of annual fee |
St.27 status event code: A-4-4-U10-U11-oth-PR1001 Fee payment year number: 9 |
|
PR1001 | Payment of annual fee |
St.27 status event code: A-4-4-U10-U11-oth-PR1001 Fee payment year number: 10 |
|
FPAY | Annual fee payment |
Payment date: 20050628 Year of fee payment: 11 |
|
PR1001 | Payment of annual fee |
St.27 status event code: A-4-4-U10-U11-oth-PR1001 Fee payment year number: 11 |
|
PN2301 | Change of applicant |
St.27 status event code: A-5-5-R10-R11-asn-PN2301 |
|
PN2301 | Change of applicant |
St.27 status event code: A-5-5-R10-R14-asn-PN2301 |
|
PN2301 | Change of applicant |
St.27 status event code: A-5-5-R10-R11-asn-PN2301 |
|
PN2301 | Change of applicant |
St.27 status event code: A-5-5-R10-R14-asn-PN2301 |
|
LAPS | Lapse due to unpaid annual fee | ||
PC1903 | Unpaid annual fee |
St.27 status event code: A-4-4-U10-U13-oth-PC1903 Not in force date: 20060810 Payment event data comment text: Termination Category : DEFAULT_OF_REGISTRATION_FEE |
|
PC1903 | Unpaid annual fee |
St.27 status event code: N-4-6-H10-H13-oth-PC1903 Ip right cessation event data comment text: Termination Category : DEFAULT_OF_REGISTRATION_FEE Not in force date: 20060810 |
|
P22-X000 | Classification modified |
St.27 status event code: A-4-4-P10-P22-nap-X000 |