JPH0630446B2 - Encoding line processing method - Google Patents
Encoding line processing methodInfo
- Publication number
- JPH0630446B2 JPH0630446B2 JP58185552A JP18555283A JPH0630446B2 JP H0630446 B2 JPH0630446 B2 JP H0630446B2 JP 58185552 A JP58185552 A JP 58185552A JP 18555283 A JP18555283 A JP 18555283A JP H0630446 B2 JPH0630446 B2 JP H0630446B2
- Authority
- JP
- Japan
- Prior art keywords
- output
- quantizer
- bits
- bit
- encoder
- 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 - Lifetime
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
- H03M7/30—Compression; Expansion; Suppression of unnecessary data, e.g. redundancy reduction
- H03M7/3002—Conversion to or from differential modulation
- H03M7/3044—Conversion to or from differential modulation with several bits only, i.e. the difference between successive samples being coded by more than one bit, e.g. differential pulse code modulation [DPCM]
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Compression, Expansion, Code Conversion, And Decoders (AREA)
Description
【発明の詳細な説明】 (a)．発明の技術分野 本発明はＴＶ映像信号の符号器に於ける符号化ライン処
理方法に係り、特に予測符号器の量子化ビット数の非整
数化を実現する方法に関するものである。DETAILED DESCRIPTION OF THE INVENTION (a). TECHNICAL FIELD OF THE INVENTION The present invention relates to a coding line processing method in a TV video signal encoder, and more particularly to a method for realizing a non-integer number of quantization bits in a predictive encoder.
(b)．従来技術と問題点 従来のＴＶ映像信号の走査ラインを符号化する符号器に
於いては、符号化出力が例えば3.5ビット等の非整数ビ
ットである量子化器を使用する必要がある場合、複数画
素（例えば２画素）毎のサンプル（11×11＝121レベル
のサンプル）を纒めて整数化した（１画素毎には3.5ビ
ットで，２3.5＝11レベルであるが７ビット／２Ｗであ
る）ワードＷとして伝送路に送出していた。(b). 2. Description of the Related Art Conventional encoders and encoders for encoding scan lines of TV video signals require a plurality of quantizers whose encoded outputs are non-integer bits such as 3.5 bits. Samples (11 × 11 = 121 level samples) for each pixel (for example, 2 pixels) are recorded and converted into integers (3.5 bits per pixel, 2 3.5 = 11 levels, but 7 bits / 2W). ) It was sent to the transmission line as word W.
一例として、伝送路約32.064ＭＨｚ、映像標本化周波数
9.16MHzの場合を採り上げて説明する。As an example, the transmission line is about 32.064MHz, the video sampling frequency
Take the case of 9.16MHz and explain.
伝送路はＰＣＭのハイアラキになっており、映像標本化
周波数(9.16MHz）は映像信号（4.2MHz）の２倍より少し
高い値をとっている。The transmission line is PCM high-araki, and the video sampling frequency (9.16MHz) is slightly higher than twice the video signal (4.2MHz).
此の時一標本化期間に割り当てられる量子化数は、3.5
ビット（32.064÷9.16＝3.5）となる。従って信号を伝
送するに当たり、映像信号を3.5ビット（非整数）の伝
送用符号に変換して伝送する必要がある。At this time, the number of quantizations assigned to one sampling period is 3.5.
It becomes a bit (32.064 ÷ 9.16 = 3.5). Therefore, when transmitting a signal, it is necessary to convert the video signal into a 3.5-bit (non-integer) transmission code for transmission.
第１図は此の様な場合に於ける従来技術による量子化器
の概念図を示すもので、図中Ｑは3.5ビットの量子化
器、Ｐは予測符号器、＋印は加算器、−印は減算器、Ｄ
は遅延回路、Ｄ／Ｄは符号変換器である。FIG. 1 is a conceptual diagram of a conventional quantizer in such a case, where Q is a 3.5-bit quantizer, P is a predictive encoder, + is an adder, and − Mark is subtractor, D
Is a delay circuit and D / D is a code converter.
第１図に於いて、入力信号Ａは減算器（−印）で予測符
号器Ｐ出力の予測信号Ｂとの差Δを求め、此の差分Δを
量子化器Ｑで第３図に示す如く、出力として量子化イン
デックス（点数）と量子化代表値（実線）とを送出す
る。In FIG. 1, a difference Δ between the input signal A and the prediction signal B of the prediction encoder P output is calculated by a subtractor (-mark), and this difference Δ is calculated by a quantizer Q as shown in FIG. , Quantization index (point) and quantized representative value (solid line) are sent as output.
一方加算器（＋印）では、予測符号器Ｐの出力の予測信
号Ｂと入力信号Ａとの差分Δゆ量子化器Ｑで量子化した
出力Δ′と前記予測信号Ｂとの和Ｂ＋Δ′を取り、予測
符号器Ｐでは、此の和Ｂ＋Δ′に基づいて新しい予測信
号Ｂを生成し出力する従来は第１図に示す様に3.5ビッ
ト／／Ｗに対しては、実際は16レベル（２の４乗）であ
るが其の中の11レベル（２の3.5乗）を利用する量子化
器Ｑを準備し符号化後、遅延回路Ｄと符号変換器Ｄ／Ｄ
を使用して２ワード集まって121レベル（11×11＝121）
としてから７ビット／２Ｗの符号を作り伝送路に送出す
ると云う方法を採っていた。此の為、符号器の他に第１
図で示す様な、符号変換の為の余分な回路Ｄ，Ｄ／Ｄを
必要とするという欠点があった。On the other hand, in the adder (+ mark), the difference B between the prediction signal B of the output of the prediction encoder P and the input signal A and the sum B + Δ 'of the output Δ'quantized by the quantizer Q and the prediction signal B are obtained. In the prediction encoder P, a new prediction signal B is generated and output based on this sum B + Δ '. Conventionally, as shown in FIG. 1, for 3.5 bits // W, actually 16 levels (2 (4th power), but a quantizer Q using 11 levels (2nd power of 3.5) is prepared and coded, and then a delay circuit D and a code converter D / D are used.
121 levels with 2 words gathered using (11 × 11 = 121)
Then, a method of forming a 7-bit / 2W code and transmitting it to the transmission line was adopted. For this reason, in addition to the encoder, the first
There is a drawback that extra circuits D and D / D for code conversion are required as shown in the figure.
(c)．発明の目的 本発明の目的は従来技術の有する上記の欠点を除去し、
従来よりも簡単な回路で出力が非整数ビットの量子化器
を実現できるＴＶ映像信号の符号化ライン処理方法を提
供することである。(c). OBJECT OF THE INVENTION The object of the invention is to eliminate the above-mentioned drawbacks of the prior art,
It is an object of the present invention to provide a coding line processing method for a TV video signal, which can realize a quantizer having a non-integer bit output with a simpler circuit than before.
(d)．発明の構成 上記の目的は本発明によれば、入力信号(A)と予測符号
器(P)の出力(B)との差（Δ）を量子化器(Q)により量子
化し出力として複数レベルの量子化インデックスを送出
すると共に該量子化器(Q)の出力の量子化代表値
（Δ′）と該予測符号器(P)の出力(B)とを加算し其の加
算結果（B+Δ′）を該予測符号器(P)に入力する様にし
た符号器に於いて、その出力が整数ビット相当で該ビッ
ト数が異なる量子化器を複数個(Q1,Q2)設置し、予め定
められた順序に従って前記複数個の量子化器(Q1,Q2)の
出力を切り替えて該予測符号器(P)の出力(B)との加算に
使用することにより、等価的に出力が非整数ビットの量
子化器を実現するように構成することにより達成され
る。(d). According to the present invention, the above object is obtained by quantizing the difference (Δ) between the input signal (A) and the output (B) of the predictive encoder (P) by the quantizer (Q) to obtain a plurality of levels as outputs. Of the quantization index of (Q) and the quantization representative value (Δ ′) of the output of the quantizer (Q) and the output (B) of the predictive encoder (P) are added, and the addition result (B + Δ ′) is input to the predictive encoder (P), and a plurality of quantizers (Q 1 , Q 2 ) are installed whose outputs correspond to integer bits and differ in the number of bits. , Equivalently by switching the outputs of the plurality of quantizers (Q 1 , Q 2 ) according to a predetermined order and using for addition with the output (B) of the predictive encoder (P), This is achieved by configuring the output to implement a quantizer with non-integer bits.
(e)．発明の実施例 本発明では出力が整数ビットで表現される整数形の量子
化器を複数個用意し、使用頻度を変化させることにより
目的を達成しようとするものであり、以下前記の例を使
用して説明する。(e). Embodiments of the present invention are intended to achieve the object by preparing a plurality of integer-type quantizers whose outputs are expressed by integer bits and changing the frequency of use. And explain.
量子化器として、16レベル（４ビット相当））及び８レ
ベル（（３ビット相当）の２種類を用意し、交互に切り
替えて使用する場合、等価的に3.5ビット相当の量子化
器が実現できる。Two types of quantizers, 16 levels (corresponding to 4 bits) and 8 levels (corresponding to 3 bits), are prepared and equivalently, a quantizer corresponding to 3.5 bits can be realized. .
叉或る区間は４ビット／Ｗで符号化し、他の区間では３
ビット／Ｗで符号化しても良い。即ち、４ビット／／Ｗ
の量子化器(Q1)と３ビット／Ｗの量子化器(Q2)とを交互
に１画素単位で切り換えるだけの方法ではなく、例えば
１ラインが58画素である場合は、画面の中央部分の290
画素を４ビット／Ｗの量子化器(Q1)で符号化し、両端の
145画素づつを３ビット／Ｗの量子化器(Q2)で符号化し
た場合でも、結果として等価的に非整数の3.5ビット／
Ｗの量子化器を実現することが出来る。For example, one section is coded with 4 bits / W, and other sections are coded with 3 bits.
You may encode by bit / W. That is, 4 bits // W
The quantizer (Q 1 ) and the quantizer (Q 2 ) of 3 bits / W are not only switched alternately on a pixel-by-pixel basis. For example, if one line consists of 58 pixels, Part of 290
Pixels are encoded with a 4-bit / W quantizer (Q 1 ) and
Even when 145 pixels are encoded by a 3-bit / W quantizer (Q 2 ), the result is equivalently non-integer 3.5 bits /
A W quantizer can be realized.
第２図は本発明の一実施例を示すもので、図中Ｑ１、Ｑ
２は夫々４ビット、３ビットの量子化器、Ｐは予測符号
器、ＳＥＬは選択回路、＋印は加算器、−印は減算器を
表すもとする。FIG. 2 shows an embodiment of the present invention, in which Q1 and Q in FIG.
2 is a 4-bit quantizer, 3 is a 3-bit quantizer, P is a predictive encoder, SEL is a selection circuit, + is an adder, and-is a subtractor.
以下第２図に従って本発明の詳細を説明する。The present invention will be described in detail below with reference to FIG.
最初に、選択回路ＳＥＬは、４ビット／Ｗの量子化器Q1
の出力を選択し使用する。入力の符号化されるべきＴＶ
映像信号は、第１図の場合と同様に、先ず４ビット／Ｗ
の量子化器Q1により２４＝16レベルの符号に変換される
が、次に選択回路ＳＥＬは、３ビット／／Ｗの量子化器
Q2の出力を選択し、入力のＴＶ映像信号を３ビット／Ｗ
の量子化器Q2により２３＝８レベルの符号に変換する。
そして次は再び４ビット／Ｗの量子化器Q1の出力が選択
されて順次送出される。此の様に、入力の符号化される
べきＴＶ映像信号が、４ビット符号の信号と３ビット符
号の信号とに交互に変換されて出力されることにより、
実効的に非整数の3.5ビット符号の信号に変換され出力
されることになる。First, the selection circuit SEL is a 4-bit / W quantizer Q 1
Select and use the output of. TV to be encoded in the input
As for the video signal, as in the case of FIG. 1, first, 4 bits / W
Is converted into a code of 2 4 = 16 levels by the quantizer Q 1 of, and the selection circuit SEL is a 3 bit // W quantizer.
Q 2 output is selected and input TV video signal is 3 bits / W
Quantizer Q 2 converts the code into 2 3 = 8 level code.
Then, next, the output of the quantizer Q 1 of 4 bits / W is again selected and sequentially transmitted. Thus, the input TV video signal to be encoded is alternately converted into a 4-bit code signal and a 3-bit code signal and output,
It is effectively converted to a non-integer 3.5-bit code signal and output.
更に本発明によると、第２図に於いて点線で囲された部
分はＲＯＭ一つで実現することが出来るので、従来の回
路に比較して回路が大変簡単となる。Further, according to the present invention, the portion surrounded by the dotted line in FIG. 2 can be realized by one ROM, so that the circuit becomes very simple as compared with the conventional circuit.
(f)．発明の効果 以上詳細に説明した様に本発明によれば、従来よりも簡
単な回路でＴＶ映像信号に対して非整数ビットを出力す
る量子化器を実現出来ると云う大きな効果が有る。(f). Effects of the Invention As described in detail above, according to the present invention, there is a great effect that a quantizer which outputs a non-integer bit to a TV video signal can be realized by a circuit simpler than the conventional one.
第１図は従来技術による量子化器の概念図を示すもの
で、図中Ｑは3.5ビットの量子化器、Ｐは予測符号器、
＋印は加算器、−印は減算器、Ｄは遅延回路、Ｄ／Ｄは
符号変換器である。 ・第２図は本発明の一実施例を示すもので、図中Ｑ１、
Ｑ２は夫々４ビット、３ビットの量子化器、Ｐは予測符
号器、ＳＥＬは選択回路、＋印は加算器、−印は減算器
を表わすものとする。 第３図は、上記の量子化器Ｑ，Ｑ１，Ｑ２の入力と出力
の一般的な説明図であり、該量子化器の入力は、−印の
減算器の出力の予測誤差信号Δであり、該量子化器の出
力は、入力の予測誤差信号Δのレベルを識別し非整数の
3.5ビットや整数の３ビット又は４ビット等で表現され
る複数レベルに量子化された出力であり、量子化インデ
ックス（点数）と量子化代表値（実線）とが送出され
る。そして、＋印の加算器へは、前記量子化代表値（実
線）のΔ′が送出される。FIG. 1 is a conceptual diagram of a quantizer according to the prior art, in which Q is a 3.5-bit quantizer, P is a predictive encoder,
The + symbol is an adder, the − symbol is a subtractor, D is a delay circuit, and D / D is a code converter. FIG. 2 shows an embodiment of the present invention, in which Q1,
It is assumed that Q2 is a 4-bit quantizer, 3-bit quantizer, P is a predictive encoder, SEL is a selection circuit, + is an adder, and-is a subtractor. FIG. 3 is a general explanatory view of the inputs and outputs of the above-mentioned quantizers Q, Q1, Q2. The input of the quantizer is the prediction error signal Δ of the output of the subtractor indicated by −. , The output of the quantizer identifies the level of the input prediction error signal Δ,
The output is quantized into a plurality of levels represented by 3.5 bits or an integer of 3 bits or 4 bits, and the quantization index (point) and the quantized representative value (solid line) are transmitted. Then, the quantized representative value (solid line) Δ'is sent to the adder indicated by +.
フロントページの続き (72)発明者 平岡 誠 神奈川県川崎市中原区上小田中1015番地 富士通株式会社内 (72)発明者 福田 裕 神奈川県川崎市中原区上小田中1015番地 富士通株式会社内 (72)発明者 黒田 英夫 神奈川県横須賀市武１丁目2356番地 日本 電信電話公社横須賀電気通信研究所内 (72)発明者 武川 直樹 神奈川県横須賀市武１丁目2356番地 日本 電信電話公社横須賀電気通信研究所内Front Page Continuation (72) Inventor Makoto Hiraoka 1015 Kamiodanaka, Nakahara-ku, Kawasaki, Kanagawa, Fujitsu Limited (72) Inventor Yu Fukuda 1015 Uedanaka, Nakahara-ku, Kawasaki, Kanagawa (72) Invention Person Hideo Kuroda 1, 2356 Take, Yokosuka City, Kanagawa Prefecture Yokosuka Electro-Communications Research Laboratories, Nippon Telegraph and Telephone Corporation (72) Inventor Naoki Takekawa 1, 2356 Take, Yokosuka City, Kanagawa Pref.
Claims (1)
の差（Δ）を量子化器(Q)により量子化し出力として複
数レベルの量子化インデックスを送出すると共に該量子
化器(Q)の出力の量子化代表値（Δ′）と該予測符号器
(P)の出力(B)とを加算し其の加算結果（B+Δ′）を該予
測符号器(P)に入力する様にした符号器に於いて、 その出力が整数ビット相当で該ビット数が異なる量子化
器を複数個(Q1,Q2)設置し、予め定められた順序に従っ
て前記複数個の量子化器(Q1,Q2)の出力を切り替えて該
予測符号器(P)の出力(B)との加算に使用することによ
り、等価的に出力が非整数ビットの量子化器を実現する
ことを特徴とする符号化ライン処理方法。1. A quantizer (Q) quantizes a difference (Δ) between an input signal (A) and an output (B) of a predictive encoder (P), and outputs a multi-level quantization index as an output. The quantized representative value (Δ ′) of the output of the quantizer (Q) and the predictive encoder
In an encoder that adds the output (B) of (P) and the addition result (B + Δ ′) to the predictive encoder (P), the output is equivalent to an integer bit. A plurality of quantizers with different numbers of bits (Q 1 , Q 2 ) are installed, and the predictive encoder by switching the outputs of the plurality of quantizers (Q 1 , Q 2 ) according to a predetermined order ( A coding line processing method, which is equivalent to realizing a quantizer having a non-integer bit output by being used for addition with an output (B) of (P).
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP58185552A JPH0630446B2 (en) | 1983-10-04 | 1983-10-04 | Encoding line processing method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP58185552A JPH0630446B2 (en) | 1983-10-04 | 1983-10-04 | Encoding line processing method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| JPS6076816A JPS6076816A (en) | 1985-05-01 |
| JPH0630446B2 true JPH0630446B2 (en) | 1994-04-20 |
Family
ID=16172801
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| JP58185552A Expired - Lifetime JPH0630446B2 (en) | 1983-10-04 | 1983-10-04 | Encoding line processing method |
Country Status (1)
| Country | Link |
|---|---|
| JP (1) | JPH0630446B2 (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH065888B2 (en) * | 1986-02-14 | 1994-01-19 | 富士写真フイルム株式会社 | Image data quantization method and apparatus |
-
1983
- 1983-10-04 JP JP58185552A patent/JPH0630446B2/en not_active Expired - Lifetime
Also Published As
| Publication number | Publication date |
|---|---|
| JPS6076816A (en) | 1985-05-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| FI98421C (en) | Adaptive quantization encoder / decoder | |
| US4922510A (en) | Method and means for variable length coding | |
| JPS61242481A (en) | How to organize image data of digital TV signal | |
| EP0439624A1 (en) | Control system for encoding image | |
| US5115241A (en) | Predictive coding device with increased resolution | |
| EP1853068A1 (en) | Processing of images using a limited number of bits | |
| US4639778A (en) | Embedding quantization system for vector signals | |
| JP3016456B2 (en) | Adaptive variable length coding method | |
| US5043809A (en) | Encoding apparatus | |
| JPS6342988B2 (en) | ||
| US4979187A (en) | Predictive coding device | |
| JPH0630446B2 (en) | Encoding line processing method | |
| EP0079658B1 (en) | Differential pulse code modulation transmission system | |
| JPH0388488A (en) | Picture coding system | |
| JPH0310486A (en) | Video encoding device | |
| JP2548122B2 (en) | Encoder | |
| SU1056455A1 (en) | Differential pulse coder of television signal | |
| ONO | Arithmetic Codes for Image Coding: Their Advantages and the Effort to Make Them Practical | |
| JPS6028453B2 (en) | Encoding method | |
| FI83714B (en) | PROCEDURE FOR PREPARATION OF PROCEDURES | |
| JPS63133764A (en) | Encoder | |
| JP2547479B2 (en) | Image coding control system | |
| JP3168444B2 (en) | Quantizer for predictive coding | |
| JP2633143B2 (en) | Predictive decoding device | |
| Krishnasami et al. | Low-bit-rate encoder for picture signals using a centre-clipping quantiser |