JPH0392977A - Method and device for eliminating noise of binary picture - Google Patents
Method and device for eliminating noise of binary pictureInfo
- Publication number
- JPH0392977A JPH0392977A JP1231291A JP23129189A JPH0392977A JP H0392977 A JPH0392977 A JP H0392977A JP 1231291 A JP1231291 A JP 1231291A JP 23129189 A JP23129189 A JP 23129189A JP H0392977 A JPH0392977 A JP H0392977A
- Authority
- JP
- Japan
- Prior art keywords
- removal
- candidate
- adjacent
- continuous
- noise
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims description 17
- 238000004458 analytical method Methods 0.000 claims description 8
- 239000000284 extract Substances 0.000 abstract description 7
- 230000008030 elimination Effects 0.000 abstract 13
- 238000003379 elimination reaction Methods 0.000 abstract 13
- 238000000605 extraction Methods 0.000 description 9
- 238000010586 diagram Methods 0.000 description 2
- 239000000470 constituent Substances 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
Landscapes
- Facsimile Image Signal Circuits (AREA)
- 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゛１″と゛Ｏ″等の２値によって表現される２
値画像からノイズを除去して鼾明で美麗な２値画像を得
る２値画像のノイズ除去方法および装置に関する。[Detailed Description of the Invention] [Industrial Application Field] The present invention relates to 2 expressed by binary values such as ``1'' and ``O''.
The present invention relates to a binary image noise removal method and apparatus for removing noise from a value image to obtain a clear and beautiful binary image.
一般に、各種の情報処理装置においては、処理を施した
情報を文字、図形等を介してディスプレー装置や用紙に
出力するようにしている。Generally, various information processing devices output processed information to a display device or paper via characters, graphics, etc.
これらの文字や図形等は、複数の画素からなる画像フィ
ールド上に、各画素がその文字や図形の構成画素である
か否かを“１”と゛Ｏ”の２つの数値をもって表現され
て、ディジタル信号処理が可能な状態として取扱われて
いる。These characters, figures, etc. are expressed digitally on an image field consisting of multiple pixels, using two numbers, "1" and "O", to indicate whether each pixel is a constituent pixel of that character or figure. It is treated as a state in which signal processing is possible.
このような２値画像となる文字や図形の利用態様は２つ
に大別される。その一方においては、各文字についてそ
れぞれ画像フィールド上の配置を予め設定しておいて、
各種の情報の出力媒体として利用している。The usage of characters and graphics that become such binary images can be roughly divided into two types. On the other hand, the arrangement of each character on the image field is set in advance,
It is used as an output medium for various information.
他方においては、必要に応じて、手寂き文字等をイメー
ジリーダで読取り、その読取った２値画像からなる文字
等を出力媒体として利用している。On the other hand, if necessary, handwritten characters and the like are read by an image reader, and the characters and the like made up of the read binary images are used as an output medium.
また、読取った２値画像としての文字等の形状のバラン
スを良いものとするため、不必要な画素を除去したり、
必要な画素を加入したりしてノイズを除去することによ
り、バランスの良い文字等からなる２ｆｆｉ画像を作成
している。In addition, in order to improve the balance of the shapes of characters, etc. as read binary images, unnecessary pixels are removed,
By adding necessary pixels and removing noise, a 2ffi image consisting of well-balanced characters, etc. is created.
〔発明が解決しようとする課題）
前述したノイズの除去を行なうために、従来は、ノイズ
除去すべき２値画像をＣＲＴ上に映し出し、それを人が
目視することによりバランスの悪い箇所を発見し、バラ
ンスを良くするようにノイズを除去して、当該２値画像
のメモリを辺直し、バランスの良い２ ＋１画像を得る
ようにしている。[Problem to be Solved by the Invention] In order to remove the noise described above, conventionally, a binary image to be noise-removed is projected on a CRT, and a person visually observes the image to discover unbalanced areas. , noise is removed to improve the balance, and the memory of the binary image is edited to obtain a well-balanced 2+1 image.
しかしながら、前記従来方法によれば、各２値画像毎に
人手によりノイズ除去操作を行なわなｔノればならず、
またノイズ除去の基準操作者毎に相違することもあり、
良好なノイズ除去を行なうことができなかった。However, according to the conventional method, the noise removal operation must be performed manually for each binary image.
Also, the standards for noise removal may differ depending on the operator.
It was not possible to perform good noise removal.
本発明はこれらの点に誂みてなされたものであり、ノイ
ズ除去対象となる２値画像の構成に最適なノイズ除去を
自動的に行なうことができ、２値画像の全体に対して適
格なノイズ除去を施したバランスの良い２値画像を得る
ことができる２値画像のノイズ除去方法および装置を提
供することを目的とする。The present invention has been made with these points in mind, and can automatically perform noise removal that is optimal for the configuration of the binary image that is the target of noise removal, and removes noise that is suitable for the entire binary image. It is an object of the present invention to provide a method and apparatus for removing noise from a binary image, which can obtain a well-balanced binary image.
本発明の２値画像のノイズ除去方法は、２値画像を所定
方向に沿って走査し、ノイズとして除去すべき値の前記
走査方向の連続画素数が除去基準画素数以下の部分を除
去候補として抽出し、前記除去候補の前記走査方向と直
交方向に隣接する各隣接走査線上の前記除去候補と連続
する隣接連続画素数をそれぞれ抽出し、前記各隣接走査
線に前記除去候補と連続する画素が共に存在しない場合
と、一方の隣接走査線のみに前記除去候補と連続づ−る
画索が存在し、かつ、前記除去候補の除去基準画素数に
対する隣接連続画素数の比率が除去基準比率以下の場合
とに、それぞれ前記除去候補をノイズとして除去するこ
とを特徴とする。The noise removal method for a binary image of the present invention scans a binary image along a predetermined direction, and selects as a removal candidate a portion where the number of consecutive pixels in the scanning direction of values to be removed as noise is equal to or less than the removal reference pixel number. and extracting the number of consecutive adjacent pixels that are continuous with the removal candidate on each adjacent scanning line adjacent to the scanning direction of the removal candidate in a direction perpendicular to the scanning direction, and the number of adjacent continuous pixels that are continuous with the removal candidate on each adjacent scanning line are extracted. There are two cases in which an image line continuous with the removal candidate exists in only one adjacent scanning line, and the ratio of the number of adjacent continuous pixels to the removal reference pixel number of the removal candidate is less than or equal to the removal reference ratio. In each case, the removal candidates are removed as noise.
本発明の２値画像のノイズ除去装置は、２ｉｆｉ画像を
所定方向に沿って走査し、ノイズとして除去すべき値の
前記走査方向の連続画素数が除去基準画素数以下の部分
を除去候補として抽出する除去候′４抽出手段と、前記
除去候補の前記走査方向と直交方向に隣接する各隣接走
査線上の前記除去候補と連続する隣接連続画素数をそれ
ぞれ抽出する隣接連続画素数抽出手段と、前記各隣接走
査線に前記除去候補と連続する画素が共に存在しない場
合と、一方の隣接走査線のみに前記除去候捕と連続する
画素が存在し、かつ、前記除去候補の除去基準画素数に
対する隣接連続画素数の比率が除去Ｍ準比率以下の場合
とに、それぞれ前記除去候補をノイズとして除去する除
去分析手段とを有づることを特徴とする。The binary image noise removal device of the present invention scans a 2ifi image along a predetermined direction, and extracts as a removal candidate a portion where the number of consecutive pixels in the scanning direction of values to be removed as noise is equal to or less than the removal reference pixel number. a removal candidate '4 extraction means for extracting the number of adjacent continuous pixels that are continuous with the removal candidate on each adjacent scanning line adjacent to the removal candidate in a direction orthogonal to the scanning direction; There are cases in which there are no pixels continuous with the removal candidate in each adjacent scanning line, and cases in which pixels continuous with the removal candidate exist in only one adjacent scanning line, and the pixels are adjacent to the removal reference pixel number of the removal candidate. The present invention is characterized in that it includes a removal analysis means that removes the removal candidates as noise when the ratio of the number of consecutive pixels is equal to or less than the removal M quasi-ratio.
本発明によれば、本発明方法に従って本発明装置を動作
させることにより、２ｆｔＪ画像に対する良好なノイズ
除去を施すことができる。According to the present invention, by operating the apparatus of the present invention according to the method of the present invention, it is possible to perform good noise removal on a 2ftJ image.
すなわち、先ず除去候補抽出手段が２値画像を所定方向
例えば横方向に走査し、ノイズとして除去すべき値例え
ば“１”の前記横走査方向への連続する画素数を求め、
更にその連続ｉｉ！ｉ素数が除去基準画素数以下の部分
を除去候補として抽出する。That is, first, the removal candidate extracting means scans the binary image in a predetermined direction, for example, in the horizontal direction, and calculates the number of consecutive pixels in the horizontal scanning direction that have a value, for example, "1", which should be removed as noise.
Furthermore, that series ii! A portion where the i prime number is less than or equal to the removal reference pixel number is extracted as a removal candidate.
次に、隣接連続画木Ｗ１ｌＩＩｌ出手段が、求められた
除去候補の走査方向と直交方向に隣接する各隣接走査線
例えば縦方向の両隣りの走査線上の除去候捕と連続して
いる隣接連続画素数をそれぞれ抽出する。Next, the adjacent continuous drawing tree W1lIIl outputting means generates an adjacent continuous image that is continuous with the removal candidate on each adjacent scanning line adjacent in the direction perpendicular to the scanning direction of the obtained removal candidate, for example, on both adjacent scanning lines in the vertical direction. Extract the number of pixels for each.
続いて、除去分析手段が、除去方向を除去すべきか否か
を判断し、次の条件を満す除去候補をノイズとして除去
する。すなわち、除去候補の各隣接走査線に除去候補と
連続する画素が共に存在しない場合と、一方の隣接走査
線のみに除去候補と連続する画素が存在し、かつ、除去
候補の除去基準画素数に対する隣接連続画素数の比率が
除去基準比率以下の場合に、それぞれ除去候補をノイズ
として除去する。Subsequently, the removal analysis means determines whether or not the removal direction should be removed, and removes removal candidates that satisfy the following conditions as noise. In other words, there are cases in which there are no pixels continuous with the removal candidate in each adjacent scanning line of the removal candidate, and cases in which there are pixels continuous with the removal candidate in only one adjacent scanning line, and the number of pixels that are continuous with the removal candidate is When the ratio of the number of adjacent continuous pixels is less than or equal to the removal reference ratio, each removal candidate is removed as noise.
（実施例）
以下、本允明の実施例を第１図から第３図について説明
する。(Example) Hereinafter, an example of Masaaki Moto will be described with reference to FIGS. 1 to 3.
第１図は本発明のノイズ除去装置の一実施例を小してい
る。FIG. 1 shows an embodiment of the noise removal device of the present invention on a smaller scale.
本実施例装置は、除去候補抽出千段１、隣接運続画素数
抽出手段２および除去分析千段３とこれらを関連動作さ
せるＣＰＵ４とによって形成されている。The apparatus of this embodiment is formed by a removal candidate extraction stage 1, an adjacent operation pixel number extraction means 2, a removal analysis stage 3, and a CPU 4 that operates these in conjunction with each other.
この除去候補抽出千段１は、２値画像を所定方向に沿っ
て走査し、ノイズとして除去すべき値の前記走査方向の
連続画素数が除去基準画像素数以下の部分を除去候捕と
してＪｄ＋出する。隣接連続画素数抽出千段２は、前記
除去候補の前記走査方向と直交方向に隣接する各隣接走
査線上の前記除去候補と連続する隣接連続画素数をそれ
ぞれ抽出する。除去分析手段３は、前記各隣接走査線に
前記除去候補と連続する画素が共に存在しない場合と、
一方の隣接走査線のみに前記除去候補と連続する画素が
存在し、かつ、前記除去候補の除去阜単画素数に対する
隣接連続画素数の比率が除去基準比率以下の場合とに、
それぞれ前記除去候補をノイズとして除去する。This removal candidate extraction step 1 scans a binary image along a predetermined direction, and extracts Jd+ as removal candidates by scanning a binary image along a predetermined direction, and extracting Jd+ as removal candidates where the number of consecutive pixels in the scanning direction of values to be removed as noise is less than or equal to the removal reference image pixel number. do. The adjacent continuous pixel number extraction stage 2 extracts the number of adjacent continuous pixels that are continuous with the removal candidate on each adjacent scanning line adjacent to the removal candidate in a direction orthogonal to the scanning direction. The removal analysis means 3 determines whether there is no pixel continuous with the removal candidate in each adjacent scanning line;
When there are pixels continuous to the removal candidate only in one adjacent scanning line, and the ratio of the number of adjacent continuous pixels to the number of removal single pixels of the removal candidate is less than or equal to the removal reference ratio,
The respective removal candidates are removed as noise.
次に、本実施例により、第２図に示す２値画像に対する
ノイズ除去処理を説明する。Next, noise removal processing for the binary image shown in FIG. 2 will be explained using this embodiment.
第２図はＸ方向およびＹ方向にそれぞれ２５画素．有す
る１フィールド内に、山という文字を゛１”の値で表示
した２ｆｔＫ画像を示している。なお、本実施例におい
て“１″は黒を示し、図示していない゛Ｏ″の部分は白
の地の部分を示すようになっている。Figure 2 shows 25 pixels in each of the X and Y directions. The figure shows a 2ftK image in which the character ``mountain'' is displayed with a value of ``1'' in one field.In this example, ``1'' indicates black, and the portion ``O'' (not shown) is white. It is designed to show the part of the land.
この１フィールド内に示されている２値画像の表示は次
のように行なわれる。The binary image shown in this one field is displayed as follows.
今、Ｘ方向に走査すると、Ｘ方向の各走査線についてｊ
番目の走査線上におけるＸ方向に対してｎ？ｌ目の黒の
部分（値が１）のランをＸＪ（ｎ）として表示する。Now, when scanning in the X direction, for each scanning line in the X direction, j
n? with respect to the X direction on the th scanning line? The run of the lth black part (value 1) is displayed as XJ(n).
そして、ｎ番目の黒のランの長さを ＸＲ．Ｊｌ （ｎ） ｊ で表わし、ｎｉ目の黒のランの始点を ＸＲ・ｏ （ｎ） ｊ で表わすようにする。Then, the length of the nth black run is XR. Jl (n) j The starting point of the ni black run is expressed as XR・o (n) j Let it be expressed as
例えば、４番目の走査線について例示すると、ＸＲ
（１）の長さはＸＲ４！Ｊ （１）−１となり、４
その始点はＸＲ４Ｄ（１）＝１３となる。For example, for the fourth scanning line, XR
(1) Length is XR4! J(1)-1, and 4 The starting point is XR4D(1)=13.
次に、第２図に対するノイズとして除去すべきＸ方向の
黒のランの長さすなわち除去基準画素数をＲ １−２と
して、ノイズ除去処理を行なう揚Ｓ
台を第３図を利用して説明する。Next, the length of the black run in the X direction to be removed as noise with respect to FIG. do.
先ず、ステップＳＴ１において未処理の２値画像を除去
候補抽出手段１に取出す。First, in step ST1, an unprocessed binary image is taken out to the removal candidate extraction means 1.
次に、ステップＳＴ２において除去候補抽出手段１が除
去候補の抽出を行なう。Next, in step ST2, the removal candidate extraction means 1 extracts removal candidates.
本実施例においては、 ＸＲＩＩ （ｎ）≦Ｒ，Ｊ）＝２ 」 の部分を捜すこととなる。In this example, XRII (n)≦R,J)=2 ” You will be looking for that part.
第２図の２値画像においては、 Ａ：ＸＲ４Ｊ）（１）＝１ Ｂ：ＸＲ１４Ｎ （１）＝１ Ｃ：ＸＲ１４Ｎ （３）＝１ Ｄ ：ＸＲ２０（ ３ ） ＝ １ のＡ．Ｂ．Ｃ，Ｄの４点が除去候補として抽出される。In the binary image shown in Figure 2, A:XR4J)(1)=1 B:XR14N (1)=1 C:XR14N (3)=1 D:XR20 (3) = 1 A. B. Four points C and D are extracted as removal candidates.
次に、ステップＳＴ３において、隣接連続画素数抽出手
段２により各除去候補についてＹ方向に隣接するＸ方向
走査線上に、除去候補と隣接する隣接連続画素数の存在
を求める。Next, in step ST3, the adjacent continuous pixel number extracting means 2 determines the existence of the number of adjacent continuous pixels adjacent to the removal candidate on the X direction scanning line adjacent to the Y direction for each removal candidate.
ステップＳ丁３における判断がＹＥＳとされると、ステ
ップＳＴ４に進行し、除去分析千段３にＪ３いて隣接連
続画素数は除去候補の片側だけに存在するか否か判断さ
れ、片側だけの場合はＹＥＳとなりステップＳ１５に進
行する。If the determination in step S-3 is YES, the process proceeds to step ST4, where it is determined whether or not the number of adjacent continuous pixels exists only on one side of the removal candidate in J3 at step ST3 of removal analysis. becomes YES, and the process proceeds to step S15.
このステップＳ１５においては、除去候補に対する隣接
連続画素数の比率が除去基準比率Ｐ以下か否かが判断さ
れ、除去基準比率Ｐ以下の場合には、ＹＥＳとなり、ス
テップＳ１６に進行して、当該除去候補がノイズとして
除去される。In this step S15, it is determined whether or not the ratio of the number of adjacent continuous pixels to the removal candidate is less than or equal to the removal reference ratio P. If the ratio is less than or equal to the removal reference ratio P, the answer is YES, and the process proceeds to step S16. Candidates are removed as noise.
ここで本実施例における各除去候補Ａ．Ｂ，Ｃ，Ｄにつ
いてみると、各除去候補とも、ステップＳＴ５へ進行し
、比率が除去阜準比率Ｐ以下であるか否かの判断を受け
る。Here, each removal candidate A. Regarding B, C, and D, each removal candidate proceeds to step ST5, where it is determined whether the ratio is equal to or less than the removal standard ratio P.
今、除去基準比率ｐ＝ｏ．ｉとする。Now, the removal standard ratio p=o. Let it be i.
Ａ．Ｂ．Ｃの各除去候補にａ３ける比率は、それぞれ０
．３３３となりＰ＝０．１より大きいのでＮＯと判断さ
れ、ステップＳＴ７に進行してノイズとして除去されな
い。A. B. The ratio of a3 to each removal candidate in C is 0.
．． 333, which is larger than P=0.1, so the determination is NO, and the process proceeds to step ST7, where it is not removed as noise.
Ｄの除去候補においては、隣接連続画素数がＸＲ２１ｆ
Ｊ （１）−２１
であり、比率は１／２１＝０．０４８となり、Ｐ＝０．
１より小さいのでＹＥＳと判断され、ステップＳＴ６に
進行して除去候補Ｄはノイズとして除去される。In the removal candidate of D, the number of adjacent continuous pixels is XR21f
J (1)-21, the ratio is 1/21=0.048, and P=0.
Since it is smaller than 1, it is determined as YES, and the process proceeds to step ST6, where the removal candidate D is removed as noise.
また、ステップＳＴ３において除去候袖にＹ方向に隣接
する２木の走査線上に、共に連続する画素が存在しない
場合には、Ｎｏと判断され、スデップＳ１６に進行して
当該除去候補はノイズとして除去される。Further, in step ST3, if there are no consecutive pixels on two scanning lines adjacent to the removal candidate in the Y direction, it is determined No, and the process proceeds to step S16, where the removal candidate is removed as noise. be done.
一方、ステップＳＴ４において、除去候補にＹ方向に隣
接する２本の走査線上に、共に連続する画素が存在する
場合には、Ｎｏと判断され、ステップＳＴ７に進行して
、当該除去候補はノイズではないとして２１ｉｆｌ画像
に含められたままとされる。On the other hand, in step ST4, if there are continuous pixels on two scanning lines adjacent to the removal candidate in the Y direction, it is determined No, and the process proceeds to step ST7, where the removal candidate is not noise. It is assumed that there is no such data, and it remains included in the 21ifl image.
このようにしてノイズ除去候補を施された２１直画像の
情報は、除去分析手段３より図示しないメモリに出力さ
れ記憶される。Information on the 21-direction image subjected to noise removal candidates in this manner is outputted from the removal analysis means 3 to a memory (not shown) and stored therein.
なお、前記実施例においては、黒の部分についてノイズ
処理を施したが、Ｍ Ｏ Ｐ＋の値の白の部分について
も同様にして行なうことができる。また、Ｙ方向に走査
しながら同様のノイズ除去処理を施すようにしてもよい
。In the above embodiment, the noise processing was performed on the black portion, but the same process can be applied to the white portion of the M O P+ value. Further, similar noise removal processing may be performed while scanning in the Y direction.
また、本発明【ま前記実施例に限定されるものではなく
、必要に応じて変史ずることができる。Further, the present invention is not limited to the above-mentioned embodiments, and can be modified as necessary.
本発明の２［画像のノイズ除去方法およびｉ置はこのよ
うに構戊され作用するものであるから、ノイズ除去対象
となる２値画像の構成に最適なノイズ除去を自動的に行
なうことができ、２値画像の全体に対して適格なノイズ
除去を施したバランスの良い２値画像を得ることができ
る等の効果を奏する。Since the image noise removal method and i-position of the present invention are configured and operated in this way, it is possible to automatically perform noise removal that is optimal for the configuration of the binary image that is the object of noise removal. , it is possible to obtain a well-balanced binary image by appropriately removing noise from the entire binary image.
第１図は本発明の２Ｑｆｉ画像のノイズ除去装置の一実
施例を示す概略図、第２図は２値画像の一例を示す説明
図、第３図は本発明方法を示すフローヂャートである。
１・・・除去候補抽出手段、
２・・・隣接連続画素数抽
出手段、
３・・・除去分析手段、
４・・・ｃｐｕ．
出誼人代理人
中
尾
俊
輔
第
１
図
第３図FIG. 1 is a schematic diagram showing an embodiment of the 2Qfi image noise removal device of the present invention, FIG. 2 is an explanatory diagram showing an example of a binary image, and FIG. 3 is a flowchart showing the method of the present invention. DESCRIPTION OF SYMBOLS 1... Removal candidate extraction means, 2... Adjacent continuous pixel number extraction means, 3... Removal analysis means, 4... cpu. Representative Shunsuke Nakao Figure 1 Figure 3
Claims (1)
除去すべき値の前記走査方向の連続画素数が除去基準画
素数以下の部分を除去候補として抽出し、前記除去候補
の前記走査方向と直交方向に隣接する各隣接走査線上の
前記除去候補と連続する隣接連続画素数をそれぞれ抽出
し、前記各隣接走査線に前記除去候補と連続する画素が
共に存在しない場合と、一方の隣接走査線のみに前記除
去候補と連続する画素が存在し、かつ、前記除去候補の
除去基準画素数に対する隣接連続画素数の比率が除去基
準比率以下の場合とに、それぞれ前記除去候補をノイズ
として除去することを特徴とする２値画像のノイズ除去
方法。 ２）２値画像を所定方向に沿って走査し、ノイズとして
除去すべき値の前記走査方向の連続画素数が除去基準画
素数以下の部分を除去候補として抽出する除去候補抽出
手段と、前記除去候補の前記走査方向と直交方向に隣接
する各隣接走査線上の前記除去候補と連続する隣接連続
画素数をそれぞれ抽出する隣接連続画素数抽出手段と、
前記各隣接走査線に前記除去候補と連続する画素が共に
存在しない場合と、一方の隣接走査線のみに前記除去候
補と連続する画素が存在し、かつ、前記除去候補の除去
基準画素数に対する隣接連続画素数の比率が除去基準比
率以下の場合とに、それぞれ前記除去候補をノイズとし
て除去する除去分析手段とを有することを特徴とする２
値画像のノイズ除去装置。[Scope of Claims] 1) A binary image is scanned along a predetermined direction, and a portion where the number of consecutive pixels in the scanning direction of values to be removed as noise is equal to or less than the removal reference pixel number is extracted as a removal candidate, If the number of adjacent continuous pixels continuous with the removal candidate on each adjacent scanning line adjacent to the removal candidate in a direction perpendicular to the scanning direction of the removal candidate is extracted, and there are no pixels continuous with the removal candidate on each adjacent scanning line, and when a pixel continuous with the removal candidate exists in only one adjacent scanning line, and the ratio of the number of adjacent continuous pixels to the removal reference pixel number of the removal candidate is less than or equal to the removal reference ratio, the removal is performed. A method for removing noise from a binary image, characterized by removing candidates as noise. 2) a removal candidate extracting means for scanning a binary image along a predetermined direction and extracting as removal candidates portions where the number of consecutive pixels in the scanning direction of values to be removed as noise is equal to or less than the removal reference pixel number; Adjacent continuous pixel number extracting means for extracting the number of adjacent continuous pixels that are continuous with the removal candidate on each adjacent scanning line adjacent to the candidate in a direction perpendicular to the scanning direction;
There are cases in which there are no pixels continuous with the removal candidate in each adjacent scanning line, and cases in which pixels continuous with the removal candidate exist in only one adjacent scanning line, and the number of pixels adjacent to the removal reference pixel number of the removal candidate is 2, characterized in that the method further comprises a removal analysis means for removing the removal candidate as noise when the ratio of the number of consecutive pixels is equal to or less than the removal reference ratio.
Value image noise removal device.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP1231291A JP2784059B2 (en) | 1989-09-06 | 1989-09-06 | Method and apparatus for removing noise from binary image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP1231291A JP2784059B2 (en) | 1989-09-06 | 1989-09-06 | Method and apparatus for removing noise from binary image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| JPH0392977A true JPH0392977A (en) | 1991-04-18 |
| JP2784059B2 JP2784059B2 (en) | 1998-08-06 |
Family
ID=16921307
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| JP1231291A Expired - Fee Related JP2784059B2 (en) | 1989-09-06 | 1989-09-06 | Method and apparatus for removing noise from binary image |
Country Status (1)
| Country | Link |
|---|---|
| JP (1) | JP2784059B2 (en) |
-
1989
- 1989-09-06 JP JP1231291A patent/JP2784059B2/en not_active Expired - Fee Related
Also Published As
| Publication number | Publication date |
|---|---|
| JP2784059B2 (en) | 1998-08-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP4077094B2 (en) | Color document image recognition device | |
| JP3753357B2 (en) | Character extraction method and recording medium | |
| KR0167616B1 (en) | Image processing apparatus and method of the same | |
| JP3636936B2 (en) | Grayscale image binarization method and recording medium recording grayscale image binarization program | |
| US6983071B2 (en) | Character segmentation device, character segmentation method used thereby, and program therefor | |
| JPH0392977A (en) | Method and device for eliminating noise of binary picture | |
| JPH0362177A (en) | Picture processor | |
| JPH06339019A (en) | Area separation system for document picture by discrete cosine transformation | |
| JPH07101455B2 (en) | Line detection method and apparatus | |
| JPH11341295A (en) | Image processing method | |
| JPH0793563A (en) | Image processor | |
| JP2915700B2 (en) | Image processing method and apparatus | |
| JP3030126B2 (en) | Image processing method | |
| US7373015B2 (en) | Image processing device and method | |
| JP3545227B2 (en) | Image processing method and apparatus, optical character reader | |
| JPH05300372A (en) | High speed sorting method for median filter | |
| JP3302855B2 (en) | Region extraction method and apparatus | |
| JP2001053977A (en) | Color picture contour line extracting device | |
| JP2840485B2 (en) | Image processing method and apparatus | |
| JPH07325045A (en) | Inspection and recognition apparatus for black print on color background | |
| JP3030814B2 (en) | Noise component removal method and recording medium recording noise component removal program | |
| JP3183916B2 (en) | Color extraction device | |
| JPH02166583A (en) | Character recognizing device | |
| JPS60221873A (en) | Thinning processing system of binary picture | |
| JPH0676110A (en) | Character/character line segmenting method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| LAPS | Cancellation because of no payment of annual fees |