JPH03102579A - Noise removal system for image data - Google Patents
Noise removal system for image dataInfo
- Publication number
- JPH03102579A JPH03102579A JP1240981A JP24098189A JPH03102579A JP H03102579 A JPH03102579 A JP H03102579A JP 1240981 A JP1240981 A JP 1240981A JP 24098189 A JP24098189 A JP 24098189A JP H03102579 A JPH03102579 A JP H03102579A
- Authority
- JP
- Japan
- Prior art keywords
- contour
- noise
- area
- image data
- black
- 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
Links
- 238000000034 method Methods 0.000 claims description 14
- 238000000605 extraction Methods 0.000 claims description 6
- 239000000284 extract Substances 0.000 claims description 3
- 238000010586 diagram Methods 0.000 description 7
- 238000007796 conventional method Methods 0.000 description 3
- 230000012447 hatching Effects 0.000 description 2
- 241000406668 Loxodonta cyclotis Species 0.000 description 1
- 241000519995 Stachys sylvatica Species 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
Landscapes
- Image Processing (AREA)
Abstract
Description
【発明の詳細な説明】
〔産業上の利用分野］
本発明はイメージデータの除去方式に関し、図面等を読
取ったイメージデータ中の不要なノイズを除去するイメ
ージデータのノイズ除去方式に関する。DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to an image data removal method, and more particularly to an image data noise removal method for removing unnecessary noise in image data obtained by reading a drawing or the like.
近年、図面をＳＩ算機人力して活用することが盛んに行
われてきており、特にスキャノ等で図而を読み込んだ後
、自動的に折れ線近似や文字認識等を行って田韓機の入
力とする、図面入力装置が多く用いられている。このよ
うな装圃において、スキャナ等で読み込んだどきに、木
来原画で意図していた図形のみを入力し、不ｉ５なノイ
ズを除去づ”ることは、原画品質の向上、データ量の削
減、及びその原画を細線化したときの原画の再現竹の向
上のために重要である。In recent years, it has become popular to use drawings manually using SI computers, and in particular, after reading the drawings using Scano, etc., it automatically performs polygon line approximation, character recognition, etc., and inputs them into Tanaka machines. Many drawing input devices are used. In such a field, when reading with a scanner etc., inputting only the intended figure in the original image and removing unnecessary noise will improve the quality of the original image and reduce the amount of data. , and is important for improving the reproduction of the original drawing when the original drawing is thinned.
（従来の技術）
従来、上述のようなノイズを除去するためには、原画を
イメージデータ化した後、一種のマスク演算を原画全体
に適用して、ノイズを除去するのが一般的であった。具
体的な例としては、第７図（Ａ＞に示すように原画中の
ｎｘｍの領域の周縁の画素Ｐｉ．１ Ｐ１．２・ Ｐ
１．ｍ Ｐ２ ＩＰ３． １−Ｐｎ．１，Ｐｎ．１
Ｐｎ，２−Ｐ１．ｍ，Ｐｎ，２，Ｐｎ．３・Ｐｎ，
ｍ−１ ｆｉ中ハッチングをほどこした部分）の画素か
全て白、または全て黒であれば、このｎ ｘ ｒｎの領
域の内部のＰｉ． ｊ ＜ｉ＝２ 〜 ｎ −
１ ． ｊ＝２ 〜ｍ−１） を全て白、
または黒にするというものがある、１この操作を入力し
た全画素について行えば、入力画像のノイズを除去づる
ことかできる．１
〔発明が解決しようとする課題〕
従来方式では、第７図（Ｂ）に示づ如＜ ｎ ｘ ｒｎ
の領域にハッヂングで示す如きノイズがある場合にはノ
イズを除去することが可能である．，しかし、第７図（
Ｃ）に示す如＜ｎｘｍの＠域の画素Ｐ１，１に白画素が
あるとハッチングで示すノイズを除去できない。(Prior art) Conventionally, in order to remove the above-mentioned noise, it was common to convert the original image into image data and then apply a type of mask operation to the entire original image to remove the noise. . As a specific example, as shown in FIG.
1. m P2 IP3. 1-Pn. 1, Pn. 1
Pn,2-P1. m, Pn, 2, Pn. 3・Pn,
If all the pixels in m-1 fi (the hatched area) are all white or all black, Pi. j <i=2 ~ n −
1. j=2 ~ m-1) are all white,
Alternatively, there is a technique called blackening.1 If this operation is performed on all input pixels, it is possible to remove noise from the input image. 1 [Problem to be solved by the invention] In the conventional method, as shown in FIG. 7(B), < n x rn
If there is noise as shown by hatching in the area, it is possible to remove the noise. , However, Fig. 7 (
As shown in C), if there is a white pixel in the pixel P1,1 in the @ area of <nxm, the noise shown by hatching cannot be removed.
更に、従来方式ではノイズを原画の図形に対し３
て小さい領域と定義しており、例えば図面を複写したと
きに生じるような線状のノイズは図形に比して同等の大
きさどなるので除ムできないどいう問題があった。Furthermore, in the conventional method, noise is defined as an area that is small compared to the figure in the original image.For example, linear noise that occurs when copying a drawing is the same size as the figure, so it is difficult to remove it. The problem was that it couldn't be done.
本発明は上記の点に鑑みなされたもので、変形ノイズや
線状ノイズを除去でき、ノイズ除去の柔軟性が高いイメ
ージデータのノイズ除去方式を提供することを目的とす
る。The present invention has been made in view of the above points, and an object of the present invention is to provide an image data noise removal method that can remove deformation noise and linear noise and has high noise removal flexibility.
４
原画のイメージデータから除去する，，〔作用〕
本発明においては、各輪郭の而栢及び縦横の大ぎさ等の
輪郭情報を複数のノイズ判別パターンと比較してノイズ
であるかどうかを判別するため、従来除去できなかった
変形ノイズや複写時に生じる線状ノイズを効果的に除去
できる１，〔課題を解決するための手段）
第１図は本発明方式の原理図を示す。4. Removal from the image data of the original image. [Operation] In the present invention, contour information such as the size and vertical and horizontal dimensions of each contour is compared with a plurality of noise discrimination patterns to determine whether or not it is noise. Therefore, deformation noise that could not be removed conventionally and linear noise that occurs during copying can be effectively removed. 1. [Means for Solving the Problems] FIG. 1 shows a diagram of the principle of the method of the present invention.
同図中、輪郭抽出部１は、イメージデータで表わされる
原画上の黒領域の輪郭点列を抽出する。In the figure, a contour extraction unit 1 extracts a sequence of contour points of a black area on an original image represented by image data.
輪郭情報算出部２は、抽出された輪郭点列より黒領域及
び白領域夫々の輪郭の面積及び縦横の大きさを含む輪郭
情報を棹出する１，
ノイズ判別部３は、各輪郭の輪郭情報を予め設定された
複数のノイズ判別パターン情報と比較しで幻象の輪郭が
ノイズであるかどうかを判別する３，ノイス除去部４は
、ノイズと判別された輪郭を〔実施例〕
第２図は本発明方式の一実施例のノローチャ｛〜を示す
。The contour information calculation unit 2 generates contour information including the area and vertical and horizontal size of the contours of each black area and white area from the extracted contour point sequence 1. The noise discrimination unit 3 calculates contour information of each contour. 3. The noise removal unit 4 compares the contour of the illusion with a plurality of preset noise discrimination pattern information to determine whether it is noise or not. The flowchart of one embodiment of the method of the present invention is shown.
同図中、原画上の全ての黒領域の輪郭についてのループ
（ステップ２０）内のスデップ２１にａ３いて輪郭点抽
出を行なう、，この輪郭点抽出は例えば総研出版、田村
秀行監修、「」ンピュータ画像処理入門」の第８３頁〜
第８５頁に記載の如き、周知の方法を用いている。In the same figure, contour point extraction is performed at step a3 in step 21 in the loop (step 20) for the contours of all black areas on the original image. Introduction to Image Processing, page 83~
Well known methods are used, such as those described on page 85.
この後ステップ２２で抽出した輪郭点列を周回積分づる
ことによって、輪郭点列に囲まれる領域の画素数即ち而
梢を算出する。黒領域の外側の輪郭である外部輪郭（第
３図の横線を施した画素）か、黒領域の内部に空いた穴
の周囲の輪郭である内部輪郭（第３図の縦線を施した画
素）であるかは周回積分で求めた面積の符号によって判
別できる。また輪郭点列の座標の縦横方向夫々の最大値
及び最小値から輪郭点列によって囲まれる領域の縦横の
人ささを求める３．更に上記而積と輪郭点列の個数つま
り輪郭長との比を求める。この比によって輪郭点列で囲
まれる領域が円形に近いか線状に近いかを判別でぎる。Thereafter, in step 22, the extracted contour point sequence is circularly integrated to calculate the number of pixels in the area surrounded by the contour point sequence, that is, the circumference. Either the external contour (pixels with horizontal lines in Figure 3) is the outline outside the black area, or the internal contour (pixels with vertical lines in Figure 3) is the outline around the hole inside the black area. ) can be determined by the sign of the area obtained by circular integration. Also, calculate the vertical and horizontal size of the area surrounded by the contour point sequence from the maximum and minimum values in the vertical and horizontal directions of the coordinates of the contour point sequence.3. Furthermore, the ratio between the above-mentioned product and the number of contour point sequences, that is, the contour length is determined. Based on this ratio, it is possible to determine whether the area surrounded by the contour point sequence is close to a circle or a line.
これらの而積、縦横の大きさ、比夫々の輪郭情報は輪郭
点列毎に第４図に示す如き輪郭点テーブルに登録される
。この登録時において、内部輪郭は縦横の大きさ及び面
積に輪郭の黒画素を含むため、第３図で縦線の画素に囲
まれる白の画素を求めるために以下の補正を行なう。These product, vertical and horizontal sizes, and contour information for each ratio are registered in a contour point table as shown in FIG. 4 for each contour point sequence. At the time of this registration, since the internal contour includes black pixels of the contour in its vertical and horizontal size and area, the following correction is performed to obtain the white pixels surrounded by the vertical line pixels in FIG.
縦横の大きさ一求めた縦横の大きざ−２而積一求めた而
積〈負）→輪郭長
更にスＴツブ２３で抽出した輪郭点列上の任意の黒画素
の座標及びこれに接する白画素の座標を輪郭点アーブル
に登録する。Vertical and horizontal size - calculated vertical and horizontal size - 2 multiplication - calculated multiplication (negative) -> contour length, plus the coordinates of any black pixel on the contour point sequence extracted by the stub 23 and the white bordering it Register the pixel coordinates to the contour point arvre.
抽出されたすべての輪郭についてのループ（ステップ２
４）内のステップ２５では各輪郭にノイズに対応ずるも
のがあるかどうかが判別される、，この場合、第５図に
示すパターンＮｏ． １ −・Ｎｏ． ５ Ｗをノイズ
の特徴を表わしている、，
パターンＮ０１：小さい点状のノイズを除去するだめの
データである。Loop over all extracted contours (step 2
In step 25 of 4), it is determined whether each contour has something corresponding to noise. In this case, pattern No. 4 shown in FIG. 1-・No. 5 W represents the characteristics of noise. Pattern No. 01: This is data for removing small dot-like noise.
５×５画素までの黒点を除去する。Removes black points up to 5x5 pixels.
パターンＮｏ． ２ ： １２ロツクス、文字の貼り込
み等で生じた線状のノイズを除去するデ
ータである１，このパラメータでは
横長のノイズ（！Ｉ５画素以下、横
２００画素以下）が除去される。Pattern No. 2: 12 lox, 1 is data for removing linear noise caused by pasting characters, etc. With this parameter, horizontally long noise (!I5 pixels or less, 200 pixels or less horizontally) is removed.
パターンＮｏ ３ ： Ｎｏ ２と同様だが、こちらは
縦型のノイズを除去するためのものであ
る。縦８０画素以下、横５画素以
下のノイズが除去ざれる、，
パターンＮ０４：黒部分に生じた白抜けを黒で埋め７
るだめのものである．，３×３画素
までの微小な穴を、黒画素で埋め
る。Pattern No. 3: Same as No. 2, but this pattern is for removing vertical noise. Noise of 80 pixels vertically or less and 5 pixels horizontally or less is removed.Pattern No. 04: White spots that occur in black areas are filled with black7. , a minute hole up to 3×3 pixels is filled with black pixels.
パターンＮｏ５：面積１００画索以下、面積と周長の比
が２以下の穴を黒く塗り潰すた
めのデータ、１かすれによる線状の
穴を埋めるためのものである。Pattern No. 5: Data for filling in black holes with an area of 100 strokes or less and a ratio of area to circumference of 2 or less, and for filling in linear holes due to 1 blur.
スアップ２５で対象の輪郭がノイズに対応するものど判
別された場合にはその輪郭について輪郭点アーブルを参
照し、テーブルに登録されている黒画素座標及び白画素
序標夫々で指示される画素が黒画素、白画素のままで置
換えられてないがどうかを判別する（ステップ２６）。If the contour of the object is determined to correspond to noise in step 25, the contour point arvre is referred to for that contour, and the pixels indicated by the black pixel coordinates and white pixel order registered in the table are It is determined whether the black pixel or white pixel remains unchanged and has not been replaced (step 26).
この判別は、第６図＜Ａ）に示す如（黒画素のノイズの
内側に穴が存在した場合、先に黒画素のノイズを白画素
に買換えた後穴である白画素のノイズを黒画素にＩＳ換
えようとすると同図（Ｂ）の如き白画素から置換えられ
た黒画素のノイズが残ってしまうからである。ここで置
換えがなされていない場合にのみ、ノイズに対応づ−る
輪郭が外部輪郭であるかど８
うかを判別する（ステップ２７）。This determination is made as shown in Figure 6<A) (if there is a hole inside the noise of a black pixel, first replace the noise of the black pixel with a white pixel, then replace the noise of the white pixel with a hole). This is because if you try to replace the IS with a pixel, the noise of the black pixel replaced from the white pixel as shown in the same figure (B) will remain.Only if the replacement is not done here, the contour corresponding to the noise will remain. It is determined whether 8 is an external contour (step 27).
外部輪郭である場合には幻象の輪郭について輪郭点テー
ブルに登録されている黒画素座標を開始点としてこれに
連結している黒画素を全て白画素に置換え（スアップ２
８）、内部輪郭である場合には苅象の輪郭について輪郭
点テーブルに登録されている白画素座標を開始点として
これに連結している自画素を全て黒画素に置換える（ス
テップ２９）。If it is an external contour, the black pixel coordinates registered in the contour point table for the phantom contour are used as the starting point, and all black pixels connected to this are replaced with white pixels.
8) If it is an internal contour, the white pixel coordinates registered in the contour point table for the contour of the elephant are used as the starting point, and all self-pixels connected thereto are replaced with black pixels (step 29).
つまり、ノイズ除去処理では、ま８ｌ″テーブルに登録
された黒画素、自画素の座標｛Ｑ置の画素を調べる。お
のおのの位置に輪郭抽出時と同様、黒画素、自画素が存
在すれば、その輪郭に対応する領域にはまだノイズ除去
処理が行われていないと判断し、ノイズ除去の処理を行
う。もし、その位閤に黒画素と白画素が存扛しな（Ｊれ
ば、その輪郭に対応づ−る領域にはノイズ除去の処理が
すでに行われているので、何もしないで次の輪郭点につ
いて処理を行う。このようにして、処理済か否かを判別
することで、正しくノイズ除去を行うことができる。こ
れは第６図（Ｃ）のような複９ｆｆな場合でも有効であ
る。In other words, in the noise removal process, the black pixel and the pixel at the coordinates {Q position of the black pixel and the self-pixel registered in the ``ma8l'' table are checked.If the black pixel and the self-pixel exist at each position, as in the case of contour extraction, It is determined that noise removal processing has not yet been performed on the area corresponding to the contour, and the noise removal processing is performed.If there are no black pixels and white pixels in that area (J, then Since noise removal processing has already been performed on the area corresponding to the contour, the next contour point is processed without doing anything.In this way, by determining whether the processing has been completed or not, Noise removal can be performed correctly. This is effective even in the case of double 9ff as shown in FIG. 6(C).
これによってノイズ判別パターンにおけるパターンＮＱ
１の縦横の大きさの−ト限を３とすれば第７図（Ｃ）
に示す如きノイズを除去できると共に、第７図（Ｃ）に
示すノイズも除去できる。また、パターンＮｏ． ２
， Ｎｏ． ３によって複写時に生じる線状ノイズを除
去でき、イメージデータの品質が向上ずる。As a result, the pattern NQ in the noise discrimination pattern
If the -g limit of the vertical and horizontal size of 1 is 3, then Figure 7 (C)
In addition to being able to remove the noise shown in FIG. 7, the noise shown in FIG. 7(C) can also be removed. Also, pattern no. 2
, No. 3 makes it possible to remove linear noise that occurs during copying, improving the quality of image data.
なお、上記実施例ではノイズ除去のために原画の輪郭を
抽出しているが、例えば細線化処理を行なう装置の場合
には、細線化処理で行なう輪郭抽出を本発明方式のノイ
ズ除去１１！Ｉ即にも利用でき、処理時間の増加を小ざ
くできる，．
〔発明の効果〕
上述の如く、本発明のイメージデータのノイズ除去方式
によれば、変形ノイズや線状ノイズを除去できノイズ除
去の柔軟性が高く、実用上きわめて右用である。In the above embodiment, the outline of the original image is extracted for noise removal, but for example, in the case of an apparatus that performs line thinning processing, the outline extraction performed by line thinning processing is performed in the noise removal method 11! of the present invention method. It can be used immediately and reduces the increase in processing time. [Effects of the Invention] As described above, the image data noise removal method of the present invention can remove deformation noise and linear noise, has high flexibility in noise removal, and is extremely useful in practice.
第１図は本発明方式の原理図、
第２図は本発明方式の一実施例のノローチャト、
第３図は輪郭点列を説明づるための図、第４図は輪郭点
テーブルを示す図、
第５図はノイズ判別パターンを示す図、第６図はノイズ
を説明するための図、
第７図は従来方式を説明するための図である．．図にＬ
ｌ３いて、
１は輪郭抽出部、
２は輪郭情報算出部、
３はノイズ判別部、
４はノイズ除去部、
２０−〜２９はステップ
を示す。
１１
１２
ヘ
衣
女
｛Fig. 1 is a diagram of the principle of the method of the present invention, Fig. 2 is a norow chart of an embodiment of the method of the present invention, Fig. 3 is a diagram for explaining a contour point sequence, and Fig. 4 is a diagram showing a contour point table. Fig. 5 is a diagram showing a noise discrimination pattern, Fig. 6 is a diagram to explain noise, and Fig. 7 is a diagram to explain the conventional method. ．． L in the figure
13, 1 is a contour extraction section, 2 is a contour information calculation section, 3 is a noise discrimination section, 4 is a noise removal section, and 20- to 29 are steps. 11 12 Woman wearing clothes {
Claims (1)
を抽出する輪郭抽出部（１）と、該抽出された輪郭点列
より黒領域及び白領域夫々の輪郭の面積及び縦横の大き
さを含む輪郭情報を算出する輪郭情報算出部（２）と、 各輪郭の輪郭情報を予め設定された複数のノイズ判別パ
ターン情報と比較して対象の輪郭がノイズであるかどう
かを判別するノイズ判別部（３）と、 ノイズと判別された輪郭を原画のイメージデータから除
去するノイズ除去部（４）とを有することを特徴とする
イメージデータのノイズ除去方式。[Scope of Claims] A contour extraction unit (1) that extracts a sequence of contour points of a black area on an original image represented by image data, and extracts the area and area of the contour of each black area and white area from the extracted sequence of contour points. A contour information calculation unit (2) that calculates contour information including vertical and horizontal sizes, and a contour information calculation unit (2) that calculates contour information including vertical and horizontal sizes, and compares the contour information of each contour with a plurality of preset noise discrimination pattern information to determine whether the target contour is noise. A method for removing noise from image data, comprising: a noise discriminator (3) that discriminates; and a noise remover (4) that removes contours determined to be noise from original image data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP1240981A JPH03102579A (en) | 1989-09-18 | 1989-09-18 | Noise removal system for image data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP1240981A JPH03102579A (en) | 1989-09-18 | 1989-09-18 | Noise removal system for image data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| JPH03102579A true JPH03102579A (en) | 1991-04-26 |
Family
ID=17067544
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| JP1240981A Pending JPH03102579A (en) | 1989-09-18 | 1989-09-18 | Noise removal system for image data |
Country Status (1)
| Country | Link |
|---|---|
| JP (1) | JPH03102579A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7400779B2 (en) | 2004-01-08 | 2008-07-15 | Sharp Laboratories Of America, Inc. | Enhancing the quality of decoded quantized images |
| US7424168B2 (en) | 2003-12-24 | 2008-09-09 | Sharp Laboratories Of America, Inc. | Enhancing the quality of decoded quantized images |
| US7424166B2 (en) | 2003-12-24 | 2008-09-09 | Sharp Laboratories Of America, Inc. | Enhancing the quality of decoded quantized images |
| US7440633B2 (en) | 2003-12-19 | 2008-10-21 | Sharp Laboratories Of America, Inc. | Enhancing the quality of decoded quantized images |
| CN112184732A (en) * | 2020-09-27 | 2021-01-05 | 山东炎黄工业设计有限公司 | Intelligent image processing method |
-
1989
- 1989-09-18 JP JP1240981A patent/JPH03102579A/en active Pending
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7440633B2 (en) | 2003-12-19 | 2008-10-21 | Sharp Laboratories Of America, Inc. | Enhancing the quality of decoded quantized images |
| US7424168B2 (en) | 2003-12-24 | 2008-09-09 | Sharp Laboratories Of America, Inc. | Enhancing the quality of decoded quantized images |
| US7424166B2 (en) | 2003-12-24 | 2008-09-09 | Sharp Laboratories Of America, Inc. | Enhancing the quality of decoded quantized images |
| US7907787B2 (en) | 2003-12-24 | 2011-03-15 | Sharp Laboratories Of America, Inc. | Enhancing the quality of decoded quantized images |
| US7400779B2 (en) | 2004-01-08 | 2008-07-15 | Sharp Laboratories Of America, Inc. | Enhancing the quality of decoded quantized images |
| US7787704B2 (en) | 2004-01-08 | 2010-08-31 | Sharp Laboratories Of America, Inc. | Enhancing the quality of decoded quantized images |
| CN112184732A (en) * | 2020-09-27 | 2021-01-05 | 山东炎黄工业设计有限公司 | Intelligent image processing method |
| CN112184732B (en) * | 2020-09-27 | 2022-05-24 | 佛山市三力智能设备科技有限公司 | Intelligent image processing method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20200240927A1 (en) | Dirt detection on screen | |
| JPH0877362A (en) | Image processor | |
| JPH03102579A (en) | Noise removal system for image data | |
| CN109509237B (en) | Filter processing method and device and electronic equipment | |
| JP2871601B2 (en) | Character string detecting apparatus and method | |
| CN111160290B (en) | Method for extracting palm vein effective area | |
| JP2001209808A (en) | System and method for extracting object, and storage medium with stored program for object extraction | |
| JPH05274422A (en) | Cell image processing method | |
| JP2822792B2 (en) | Image noise removal device | |
| JP3223385B2 (en) | Pattern matching device for grayscale images | |
| JPH08102858A (en) | Method of clarifying picture outline part of digital picture | |
| JP3585143B2 (en) | Character string extraction method and device | |
| JP2853510B2 (en) | Image noise removal device | |
| JPH05143774A (en) | Method for correcting inclination of character document | |
| JP2938217B2 (en) | Image processing device | |
| JP2001053977A (en) | Color picture contour line extracting device | |
| JPH09138854A (en) | Method and device for extracting area of picture | |
| JPH06325156A (en) | Method and device for character shape inspection | |
| CN116957976A (en) | Target image enhancement method, target image enhancement device, and storage medium | |
| JP3046654B2 (en) | Detecting inclination of text originals | |
| JPH01296385A (en) | Method for improving picture quality of binary picture data | |
| JPS62247473A (en) | Picture processor | |
| JPH1011589A (en) | Picture processing method | |
| JPH0830723A (en) | Half-tone dot meshing area processing method | |
| JPH08263650A (en) | Picture processing method and picture processor |