JPS59146262A - System for deciding and processing binary/multi-value picture area - Google Patents

System for deciding and processing binary/multi-value picture area

Info

Publication number
JPS59146262A
JPS59146262A JP58021322A JP2132283A JPS59146262A JP S59146262 A JPS59146262 A JP S59146262A JP 58021322 A JP58021322 A JP 58021322A JP 2132283 A JP2132283 A JP 2132283A JP S59146262 A JPS59146262 A JP S59146262A
Authority
JP
Japan
Prior art keywords
value
binary
processor
picture
arithmetic
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.)
Pending
Application number
JP58021322A
Other languages
Japanese (ja)
Inventor
Nagaharu Hamada
長晴 浜田
Yasuyuki Kojima
康行 小嶋
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Hitachi Ltd
Original Assignee
Hitachi Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Hitachi Ltd filed Critical Hitachi Ltd
Priority to JP58021322A priority Critical patent/JPS59146262A/en
Publication of JPS59146262A publication Critical patent/JPS59146262A/en
Pending legal-status Critical Current

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N1/00Scanning, transmission or reproduction of documents or the like, e.g. facsimile transmission; Details thereof
    • H04N1/40Picture signal circuits
    • H04N1/40062Discrimination between different image types, e.g. two-tone, continuous tone

Abstract

PURPOSE:To offer a processing system of high speed specification exceeding the performance of an exclusive processor by performing the discriminating processing of binary/multi-value picture area by a local vicinity arithmetic exclusive processor for multi-value picture and by an external discriminating circuit, to realize easily the discrimination of the binary/multi-value picture area. CONSTITUTION:In the figure, 20-20' and 23-23' constitute a processing image buffer 24 having hour sets of 4-bit width as an alternate buffer form. 25-25''' are processors and they are controlled to obtain a maximum and a minimum value respectively, and a picture data for 4 picture elements' share of each scanning line is applied in parallel from the image buffer. The result of arithmetic is transferred among the processors via each linkage unit respectivey, and the maximum value and the minimum value in 4X4 picture elements are outputted as an output of the processor 25''' of the 4th stage. After the maximum value is latched once in a register 26, the difference with the minimum value outputted in the next timing is used at an arithmetic circuit 27 and the result is compared with a threshold value Po for binary/multi-value decision at an arithmetic circuit 28, allowing to obtain a command signal for separating the binary/ multi-value picture area.

Description

【発明の詳細な説明】 〔発明の利用分野〕 本発明は2値/多値画像領域判定処理方式に係り、特に
ファクシミリなどにおける画像データの符号・復号化の
高速な前処理方式に関する。
DETAILED DESCRIPTION OF THE INVENTION [Field of Application of the Invention] The present invention relates to a binary/multilevel image area determination processing method, and particularly to a high-speed preprocessing method for encoding/decoding image data in facsimile and the like.

〔従来技術〕[Prior art]

ファクシミリなどのように画像データの伝送装置で、写
真等の多値画像をデジタル信号として伝送する場合に、
多値画像と2値画像とを識別して符号化すれば、符号化
効率が向上することが知られている。
When transmitting multivalued images such as photographs as digital signals using image data transmission devices such as facsimiles,
It is known that encoding efficiency can be improved if multi-level images and binary images are identified and encoded.

第1図と第2図を用いて公知の判定方法を説明する。第
1図は画面の座標定義を示し、横方向が、ファクシミリ
やテレビで用いる主走査方向であり、上下方向か副走査
方向と呼ぶ。身)1図の各1」は、1画素を意味し、太
枠で囲んた・↓画累×4両系単位で前記の2値/多呟画
像の判定をイ1う。
A known determination method will be explained using FIGS. 1 and 2. FIG. 1 shows the definition of screen coordinates, where the horizontal direction is the main scanning direction used in facsimiles and televisions, and is called the vertical direction or sub-scanning direction. 1) Each 1 in 1 figure means 1 pixel, and the above-mentioned binary/multiple images are judged in units of ↓ strokes x 4, which are surrounded by a thick frame.

第2は、この4両系X4画像の1lj−位(以十ブロッ
クと称す)を抜き出したもの−(、このンDンタ中の#
度の最大値(Dmax)及び最小値(D m−i、m 
)を求め、この濃瓜差Pか、基準と7.J:る濃度1・
。よりも大きい場合には、このフロック上2値と判定し
、Poよりも小さい場合には多値画像と1′、す定する
The second one is the one extracted from the 1lj-th block (hereinafter referred to as 10 blocks) of this 4-way x4 image (, # in this printer).
Maximum value (Dmax) and minimum value (D m-i, m
) is calculated, and this difference P is calculated based on the standard and 7. J: concentration 1・
. If it is larger than Po, it is determined that the flock is binary, and if it is smaller than Po, it is determined that it is a multivalued image.

この判定を行っに適当な処理力式は知l″yAシでいな
い。
There is no known processing power formula suitable for making this determination.

第3図は、局所近傍演算を行つに適したIi!I用プロ
セッサである(詳細は、特願昭56−94048を参照
)、、本プロセッサは、8ピッl−/ il!Ij累の
アータ経路として4組のシフトレジスタト−1で構成し
たデータユニット2、M襲用の荷里係数やテンプレート
・データなどを、記憶する8ビツト×1〔3X4バイト
のRAM(ランダムアクセスメモリ)で構成したメモリ
ユニット3、画素単位の各種演算を実行する加算器(A
LU)4及び乗算器(MU+t)からなるプロセッサニ
レメン+−(PE)6へ6′で構成したプロセッサユニ
ツ1−7、それぞれのl) r> 6の演算結果を総合
するだめの加算器(AI、(J)2段8及び9で構成し
たリンケージユニット10.2値化やクラスタリングの
ためのエバリュージョン・ユニット11及び全体の演算
結果を制御するコン1−ロールユニット12がらなって
いる。このプロセッサでは、(↑) 4個の画素に夕J
’ L、 4個のP E 6が並列に演算i′iJ能、
(2)  プロセソザユニツ1〜7とリンケージ出力ッ
l−10内の各演算処理かバイブライン処理可能な構造
になっており、6Mbpsまでの入力画像に対し、リア
ルタイム処理ができる。また、コントロールユニット1
2の各種制御レジスタの内容を沓ぎ替えることかでき、
これによっ゛C基本的な局所近傍演算をほとんど実行す
ることが可能である。しかしながら、この1.、 S 
Iの仕様を越える性能の要求は高い。
FIG. 3 shows Ii! which is suitable for performing local neighborhood calculations. This processor is an 8-pin l-/il! Data unit 2 consists of four sets of shift registers 1 as the data path for Ij, and 8-bit x 1 [3 x 4-byte RAM (random access memory) for storing coefficients, template data, etc. for M ), an adder (A
Processor units 1-7 consisting of LU) 4 and multipliers (MU+t) and multipliers (MU+t), each consisting of processor units 1-7 consisting of LU) 4 and multipliers (MU+t); AI, (J) A linkage unit 10 composed of two stages 8 and 9, an evaluation unit 11 for binarization and clustering, and a control unit 12 for controlling the overall calculation results. In this processor, (↑)
' L, 4 P E 6 can be operated in parallel,
(2) It is structured so that each arithmetic processing in the processor units 1 to 7 and the linkage output unit 10 can be processed by vibrating, and real-time processing can be performed on input images up to 6 Mbps. In addition, control unit 1
You can change the contents of various control registers in 2.
This makes it possible to perform most of the basic local neighborhood operations in C. However, this 1. , S
There is a high demand for performance that exceeds the I specifications.

〔発明の目的〕[Purpose of the invention]

本発明の1」的は、2値/多値画像領域判定を容易に実
現する処理方式を提供するにあり、また本発明の他の目
的は、前記専用プロセッサの性能を越える高速仕様の処
理方式を提供するにある。
A first object of the present invention is to provide a processing method that easily realizes binary/multi-valued image area determination, and another object of the present invention is to provide a processing method with high-speed specifications exceeding the performance of the dedicated processor. is to provide.

〔発明の概要〕[Summary of the invention]

本発明は、前記プロセッサに前処理用のイメージバツハ
アと演算回路とラッチとタイミング回路とを組合せて、
2値/多値画像領域判定処理を行う。
The present invention combines the processor with an image processor for preprocessing, an arithmetic circuit, a latch, and a timing circuit,
Performs binary/multivalued image area determination processing.

〔発明の実施例〕[Embodiments of the invention]

以■、本発明の一実施例を図面を用いで説明する。第4
図は、本実施例のブロック構成を示す。
Hereinafter, one embodiment of the present invention will be described with reference to the drawings. Fourth
The figure shows the block configuration of this embodiment.

第4図において、20〜20’、2]へ一2t’。In FIG. 4, 20 to 20', 2] to 2t'.

22〜22’ 、23〜23′は4組の4ピツ1へ幅を
持つ前記処理イメージバッファ24を構成し交代バッフ
ァ形式をとっている。各メモリは1走査線分で4ピッl
−幅であり16階調を記憶することができる。すなわち
外部からの入力画像が逐次イメージバツハア24に1走
査線ずっ4走査線分を格納する間に一方の4走査線のメ
モリを使うことができる。25〜25″は前記のプロセ
ッサであり、各々のプロセッサ25に前記イメージバツ
ハアから並列に画像データを供給する。各プロセッサに
は最大値及び最小値を求めるように制御しておき、各々
の走査線の4画素分の画像データを供給する。71個の
プロセッサ間はそれぞれリンケーシュニツ1〜を介して
、順次演算結果を転送し、4段[1のプロセッサ25−
の出力として4画素×4画素内の最大111.及び最小
値を出力させる。この最大値登一度レジスタ26にラッ
チした後、次のタイミングで出力される最小値の差を演
算回路27てとり、ぞの昂果をさらに2値/多値判定の
ためのし、きい?Uj P。と演算回路28で比較する
ことにより、2値/′多値画像領域分離の為の指令信号
が1ζ1ら才しる。この信号は次の画像データの入力で
変化するから2ノツチ29で一時記憶し、また、1プロ
ツノfi↓にその値を更新できるようにカウンタ3θを
設けた領域予約回路31を経て信号の前処理回路32に
与える。このように、本プロセッサをオ、)ずかのロジ
ックと組合せて用いることにより、通常のMSl、SS
Iを用いるより・1−)はるかに簡単かつ少ない回路で
21直/多値画像頭域判疋処理回路を実現することが1
4能−C1らQ n次のこの出力信号の使用実施例を第
5図及び第6図を用い゛C説明する。第5図い5、領域
分離により2値化がディザ化を4走査線1fiJ H4
Jに処理する。
22 to 22' and 23 to 23' constitute the processing image buffer 24 having a width of 4 sets of 4 pixels 1, and are in an alternating buffer format. Each memory has 4 pixels for one scanning line
- width and can store 16 gradations. That is, while an input image from the outside is sequentially stored in the image buffer 24 for one scanning line for four scanning lines, one memory for four scanning lines can be used. 25 to 25'' are the aforementioned processors, and each processor 25 is supplied with image data in parallel from the image processor. Each processor is controlled to obtain the maximum value and the minimum value, and each scanning Image data for 4 pixels of a line is supplied.Arithmetic results are sequentially transferred between the 71 processors via linkers 1 to 1, and processors 25-1 of 4 stages [1]
The maximum output is 111. within 4 pixels x 4 pixels. and output the minimum value. After this maximum value is latched into the register 26, the difference between the minimum values output at the next timing is taken by the arithmetic circuit 27, and this result is further used for binary/multi-value determination. Uj P. By comparing the values in the arithmetic circuit 28, a command signal for binary/'multivalued image region separation is obtained by 1ζ1. Since this signal changes with the input of the next image data, it is temporarily stored in the 2-notch 29, and is pre-processed through the area reservation circuit 31 provided with a counter 3θ so that its value can be updated at 1-pron.fi↓. to circuit 32. In this way, by using this processor in combination with several logics, ordinary MSI, SS
1-) It is possible to realize a 21-digit/multi-level image head area processing circuit with much simpler and fewer circuits than using I.
An example of the use of this output signal of order Q4 from C1 to Q will be described with reference to FIGS. 5 and 6. Figure 5. Binarization and dithering by region separation 4 scanning lines 1fiJ H4
Process to J.

1走査線分のU+戊は、第51最」一段ブロックで示し
たように、ディサしきい値と、z値化のしさい値を保持
すζ・レジスタ40及び40’、その出力を2値/′デ
ィザ指令信号によって選択するスイッテ、人力データと
しきい値との比vJ)Lを行う比較回路42、ぞの出力
を符号化回路に転送するために直並列変換するだめのシ
フ1−レジスタ43からなっている。これらからなる2
値化回路は4走査線分・111〜47まで用陥ニジ又い
る。なお、ξ)ヒツト遅延回路41ば、判定処理との同
期を合わせるだめの回路である。なお、しきい値レジス
タには、システムのイニシャライズ時所定の値に設足す
る。
As shown in the 51st one-stage block, U + 戊 for one scanning line is a dither threshold value and ζ registers 40 and 40' that hold the threshold values for z-value conversion, and their outputs are converted into binary values. /'Switch selected by the dither command signal, comparator circuit 42 that performs the ratio vJ)L between the human input data and the threshold value, and the shift register 43 that converts the output from serial to parallel in order to transfer it to the encoding circuit. It consists of 2 consisting of these
The value converting circuit is used for four scanning lines from 111 to 47. Note that the hit delay circuit 41 (ξ) is a circuit for synchronizing with the determination process. Note that a predetermined value is set in the threshold register when the system is initialized.

以上の動作を表したのが第6図であり、各々の画像入力
に対してリンケージ出力は図のように゛なつており左か
ら5番目のタイミングで、画像入力を行わず、最大値出
力動作のみを行わせ、次のタイミングで最小値を出力す
ると同時に次のブロックの画像データを入力し、次のブ
ロックの判定処理を行う。これはプロセッサ内部がパイ
プライン処理になっているためである。
Figure 6 shows the above operation, and the linkage output for each image input is arranged as shown in the figure. At the fifth timing from the left, no image input is performed and the maximum value output operation is performed. At the next timing, the minimum value is output, and at the same time, the image data of the next block is input, and the next block is judged. This is due to the pipeline processing inside the processor.

以上に説明したように、本実施例によれば前記の局所近
傍演算処理プロセッサにより、容易に2値/多値画仰領
坊゛r」1定処理が実現できるほか、前記プロセッサの
能力を次のように増力する効果がある。
As explained above, according to the present embodiment, the local neighborhood arithmetic processing processor can easily realize binary/multivalued image processing, and the capabilities of the processor can be It has the effect of increasing power.

総合能力=m体能力X 16 / 5 な才9前記のブ[lツク予約回路を変形し・たり処理タ
イミンクを変更し5、能力よりも機能を優先させること
は自由である。
Total ability = m body ability

〔発明の効果〕〔Effect of the invention〕

本発明によれば、2値/多値の領域判定を容易に行オ、
)することかできるばかりでなく、使用するプロセッサ
の能力を越えた高速仕様にも適用できる効果がある。
According to the present invention, binary/multivalued area determination can be easily performed.
), it also has the effect of being applicable to high-speed specifications that exceed the capabilities of the processor used.

【図面の簡単な説明】[Brief explanation of drawings]

第1図は、2値/多値画像の領域判定方式の前提となる
画面と座標、ブロックの関係を示す説明図、第2図は、
公知の2値/多値画像領域判定方式を示す説明図、第3
図は、本発明に用いる公知の局所近傍演算専用プロセッ
サ、第4図は、本発明の実施例のブロック図、第5図は
、本発明を適用する2値/ディザ選択回路のブロック図
、第6図は、第4図及び第5図の動作タイミング゛ナヤ
ートの1例である。 24・・・イメージバツハア、25・・・プロセッサ、
26・・・ラッチ、27・・・演算回路、28・・・演
算回路、31・・・ブロック予約量図面の浄書(内容に
変更なし) 茅2 目 P二 I)艷欅−p弧l竹 PgPo  めヒき  z4直 P<f’、のヒぎ 多硫 茅3 区 羊 5.=  鵠 、?()     25   34 第52 2イ96(/−デー信Jγiイ妄ゝ 第 Z ロ 手続補正音(方式) 1シ、・1() 長 1・1   イ〒 杉 和 夫 
)(1す・I  (’l”ハ  人  ・J、 昭(II 58−’l’ +lj:、r19[:第 2
1322  Xj力 明 5・))′、(,1,2値/
多値lI!111象饋域判定処理ガ式%式% ) : (1式:t+l  ll〕4ll11す「内 ・(己・
」、すr  1111・1、イ(〕、7メー\、抽 ]
)−・″)月 9↓ (1)  代理権を証明する誓面
(2)明細書おまひ図面の全部 /−1゛/ 11゛・ idi   ll   −′)   内  百    
                         
    −」8(1,1委任状を提出する。 (2)明細書及び図面の浄書(山谷に変更なし)。 以上 一
Figure 1 is an explanatory diagram showing the relationship between the screen, coordinates, and blocks, which are the premises of the area determination method for binary/multivalued images.
Explanatory diagram showing a known binary/multivalued image area determination method, 3rd
4 is a block diagram of an embodiment of the present invention. FIG. 5 is a block diagram of a binary/dither selection circuit to which the present invention is applied. FIG. 6 is an example of the operation timing diagram of FIGS. 4 and 5. 24...Image battle, 25...Processor,
26...Latch, 27...Arithmetic circuit, 28...Arithmetic circuit, 31...Block reservation amount drawing engraving (no change in content) Kaya 2 eyes P 2 I) 艷欅-p arcl bamboo PgPo Mehiki z4DirectP<f', Nohigi Polysulfur grass 3 Ward sheep 5. = Goose? () 25 34 No. 52 2-96 (/-Design Jγi Delusion No. Z B Procedure Correction Sound (Method) 1shi,・1() Long 1.1 I〒 Kazuo Sugi
) (1su・I ('l''ha person ・J, Akira (II 58-'l' +lj:, r19[: 2nd
1322 Xj force light 5・))′, (,1,2 value/
Multi-value lI! 111 Elephant area judgment processing Ga expression % expression %): (1 expression: t+l ll) 4ll11su'・(self・
", sr 1111.1, i (), 7 me\, draw]
)−・″) Month 9↓ (1) Oath to prove authority of attorney (2) All of the specifications and drawings /−1゛/ 11゛・idi ll −′) Within 100

-"8 (1,1 Submit a power of attorney. (2) Engraving of the specification and drawings (no changes to Yamaya).

Claims (1)

【特許請求の範囲】[Claims] 1、多値画像用の局所近傍演算専用プロセッサと、外部
の判定回路とにより2値/多値画像領域判定処理を行う
ことを特徴とする2値多値画像領域判定処理方式
1. A binary/multilevel image area determination processing method characterized by performing binary/multilevel image area determination processing using a processor dedicated to local neighborhood calculations for multilevel images and an external determination circuit.
JP58021322A 1983-02-09 1983-02-09 System for deciding and processing binary/multi-value picture area Pending JPS59146262A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP58021322A JPS59146262A (en) 1983-02-09 1983-02-09 System for deciding and processing binary/multi-value picture area

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP58021322A JPS59146262A (en) 1983-02-09 1983-02-09 System for deciding and processing binary/multi-value picture area

Publications (1)

Publication Number Publication Date
JPS59146262A true JPS59146262A (en) 1984-08-22

Family

ID=12051912

Family Applications (1)

Application Number Title Priority Date Filing Date
JP58021322A Pending JPS59146262A (en) 1983-02-09 1983-02-09 System for deciding and processing binary/multi-value picture area

Country Status (1)

Country Link
JP (1) JPS59146262A (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP0488724A2 (en) * 1990-11-30 1992-06-03 Canon Kabushiki Kaisha Ink jet recording apparatus and method
JPH04296163A (en) * 1990-12-31 1992-10-20 Gold Star Co Ltd Automatic picture/character separating apparatus for image information and method thereof
JPH06293451A (en) * 1993-02-27 1994-10-21 Murata Mach Ltd Support structure of roll sheet

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP0488724A2 (en) * 1990-11-30 1992-06-03 Canon Kabushiki Kaisha Ink jet recording apparatus and method
EP0706889A1 (en) * 1990-11-30 1996-04-17 Canon Kabushiki Kaisha Ink jet recording apparatus and method
US5745145A (en) * 1990-11-30 1998-04-28 Canon Kabushiki Kaisha Ink jet recording apparatus and method
JPH04296163A (en) * 1990-12-31 1992-10-20 Gold Star Co Ltd Automatic picture/character separating apparatus for image information and method thereof
JPH06293451A (en) * 1993-02-27 1994-10-21 Murata Mach Ltd Support structure of roll sheet

Similar Documents

Publication Publication Date Title
JPH0265371A (en) Encoding device for picture
US4710823A (en) Density conversion in image reproduction
JPS59146262A (en) System for deciding and processing binary/multi-value picture area
JP3976863B2 (en) Image processing apparatus and method
JP3809274B2 (en) Image processing apparatus and method
JPH0447470A (en) Picture processing system using neural net and picture processing device using the system
JPS6341272B2 (en)
JP2575641B2 (en) Image editing processing method
JPH03277068A (en) Color picture communication equipment
JP2966426B2 (en) Color image processing apparatus and method
JPH06105162A (en) Picture processing unit
JPH11331589A (en) Image processor and image processing method
JPS58138173A (en) Reading system of facsimile picture information
JPS6337777A (en) Compression system for image data
JPS59175280A (en) Device for binary-coding video signal
JPS63117562A (en) Image data coding system
JPH08125861A (en) Picture processor
JPH1028224A (en) Image data processor
JPH02179166A (en) Binary picture predicting device
JPH01171366A (en) Data compressing device
JPS62175037A (en) Information transmission system
JPH03277073A (en) Color picture communication equipment
JPH0391876A (en) Picture processing device
JP2000125132A (en) Picture processor and control method therefor
JPH02290367A (en) Processor for adding stereoscopic shadow