JPH0997309A - Character extracting device - Google Patents
Character extracting deviceInfo
- Publication number
- JPH0997309A JPH0997309A JP7251395A JP25139595A JPH0997309A JP H0997309 A JPH0997309 A JP H0997309A JP 7251395 A JP7251395 A JP 7251395A JP 25139595 A JP25139595 A JP 25139595A JP H0997309 A JPH0997309 A JP H0997309A
- Authority
- JP
- Japan
- Prior art keywords
- color
- character
- line
- distance
- ruled line
- 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
Landscapes
- Character Input (AREA)
Abstract
Description
【０００１】[0001]
【発明の属する技術分野】本発明は、文字抽出装置に係
り、特に罫線枠のある伝票等からの文字抽出に関する。BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a character extraction device, and more particularly to character extraction from a slip having a ruled frame.
【０００２】[0002]
【従来の技術】従来、枠などでフォーマットが指定され
た原稿を読み取る場合、フォーマットに対応した色の情
報を読取り時に除去するいわゆるドロップアウトカラー
技術が用いられている。これは光源を選択したり、フィ
ルターを用いたりすることにより不要な色情報を除去
し、別の色で形成された文字情報のみを取り出すように
したものである。2. Description of the Related Art Conventionally, when reading a document whose format is specified by a frame or the like, a so-called dropout color technology is used which removes color information corresponding to the format at the time of reading. In this, unnecessary color information is removed by selecting a light source or using a filter, and only character information formed in another color is taken out.
【０００３】しかしながら、このような技術では対象と
するドロップアウトカラーの種類が制限される上、光学
系に依存するため、耐久性、高速性などの面で問題が多
かった。However, in such a technique, there are many problems in terms of durability and high speed because the type of dropout color to be targeted is limited and it depends on the optical system.
【０００４】例えば多数枚複写方式の伝票などでは、そ
れぞれを区別するために、多彩な色を使用することが多
くなってきている。このように多彩な色で印刷された伝
票においては、罫線情報を除去し、罫線枠内の文字のみ
を読み取ろうとすると、処理が極めて複雑となるという
問題があった。For example, in a multi-copy form slip, various colors are often used in order to distinguish between them. In such a slip printed in various colors, if the ruled line information is removed and only the characters in the ruled line frame are read, the processing becomes extremely complicated.
【０００５】そこでＲＧＢ空間のクラスタリング処理に
より、分布密度の高い３点すなわち文字、用紙、模様の
点を検出、その点のうち、点（０，０，０）に最も近い
点を文字、点（０，０，０）から最も遠い点を用紙、残
りを模様とする方式が提案されている（特開平４−１３
３１９０号）。Therefore, three points having a high distribution density, that is, a point of a character, a sheet, and a pattern are detected by the clustering processing of the RGB space, and the point closest to the point (0,0,0) among the points is the character, the point ( A method has been proposed in which the furthest point from (0,0,0) is the paper and the rest is the pattern (JP-A-4-13).
3190).
【０００６】[0006]
【発明が解決しようとする課題】このような方式では、
ＲＧＢ空間のクラスタリング処理時の処理量が極めて多
く、処理に多大な時間がかかるという問題があった。SUMMARY OF THE INVENTION In such a system,
There has been a problem that the amount of processing at the time of clustering processing of the RGB space is extremely large and the processing takes a lot of time.
【０００７】本発明は、前記実情に鑑みてなされたもの
で、高速処理が可能で、多彩な色、あるいは多色印刷に
対しても適用可能な文字抽出装置を提供することを目的
とする。The present invention has been made in view of the above circumstances, and an object of the present invention is to provide a character extraction device which can perform high-speed processing and can be applied to various colors or multicolor printing.
【０００８】[0008]
【課題を解決するための手段】本発明の装置の特徴は、
罫線枠のある伝票などの書類から文字情報のみを取り出
す文字抽出装置において、背景色と罫線枠の色のみが存
在する領域を色判定領域として設定する色判定領域設定
手段と、前記色判定領域設定手段で設定された領域を読
取り、その領域からＲＧＢ空間での背景色と罫線色の分
布中心を線で結び、罫線色を示す基準線を形成する罫線
色基準線決定手段と、罫線を含む文字領域を読取り、そ
の分布と前記罫線色基準線との距離を演算する演算手段
と、前記演算手段によって算出された距離があらかじめ
決定されたしきい値よりも大きいか否かを比較する比較
手段とを含み、大きいときのみ文字情報としてとりだす
ようにしたことにある。The features of the device of the present invention are:
In a character extraction device for extracting only character information from a document such as a slip having a ruled line frame, a color determination region setting means for setting a region in which only the background color and the color of the ruled line frame exist as a color determination region, and the color determination region setting A ruled line color reference line determining unit that reads a region set by the means, connects the background color and the distribution center of the ruled line color in the RGB space with a line from the region, and forms a reference line indicating the ruled line color, and a character including the ruled line A calculation means for reading the area and calculating the distance between the distribution and the ruled line color reference line; and a comparison means for comparing whether or not the distance calculated by the calculation means is larger than a predetermined threshold value. Including that, it is to be taken out as character information only when it is large.
【０００９】[0009]
【発明の実施の形態】かかる構成によれば、あらかじめ
色認識エリアを設定してこのＲＧＢ読取り出力から背景
−罫線色基準線を作成しておき、この背景−罫線色基準
線からの距離に応じて文字情報であるか罫線情報である
かを決定するようにしているため、処理量の大幅な低減
をはかるとともに高速で信頼性の高い抽出処理が可能と
なる。また文字と罫線の明度が逆転していてもよく、ま
た、多色印刷に対しても容易に適用可能である。According to such a configuration, a color recognition area is set in advance and a background-ruled line color reference line is created from this RGB read output, and the background-ruled line color reference line is adjusted in accordance with the distance. Since it is determined whether the information is character information or ruled line information, it is possible to significantly reduce the processing amount and to perform high-speed and highly reliable extraction processing. Further, the brightness of the character and the ruled line may be reversed, and the invention can be easily applied to multicolor printing.
【００１０】また、多色印刷でも色情報を検出する領域
を印刷色毎に設定して登録しておけば、容易に抽出可能
である。Further, even in multicolor printing, if the area for detecting color information is set and registered for each printing color, it can be easily extracted.
【００１１】例えば赤と青の２色で印刷されていても、
赤と青それぞれで処理した画像を組み合わせることによ
って赤と青の両方を除去することができるため処理が極
めて簡単である。For example, even if printed in two colors of red and blue,
The processing is extremely simple because both red and blue can be removed by combining the images processed with red and blue respectively.
【００１２】次に、本発明の実施例について図面を参照
しつつ詳細に説明する。Next, embodiments of the present invention will be described in detail with reference to the drawings.
【００１３】この文字情報抽出装置は、図１に機能ブロ
ック図を示すように、認識しようとする画像をよみとる
ための画像入力部１と、背景色と罫線枠の色のみが存在
する領域を色判定領域として設定する色判定領域設定手
段と、前記色判定領域設定手段で設定された領域を読取
り、その領域からＲＧＢ空間での背景色と罫線色の分布
中心を線で結び、罫線色を示す基準線を形成する罫線色
基準線決定手段とを具備し、得られた情報を格納する色
情報格納部２Ｓと、罫線を含む文字領域を読取り、その
分布と前記罫線色基準線との距離を演算する距離演算部
２Ｃと、前記距離演算部によって算出された距離があら
かじめ決定されたしきい値よりも大きいか否かを比較す
る比較手段とを含み、大きいときのみ文字情報としてと
りだすようにした文字抽出部２と、文字抽出部で抽出さ
れた文字情報を二値化する二値化処理部３と、二値化処
理部３で形成された二値化情報を認識する文字認識部４
とを具備したことを特徴とする。As shown in the functional block diagram of FIG. 1, the character information extracting apparatus includes an image input section 1 for reading an image to be recognized, and an area in which only the background color and the color of the ruled line frame exist. The color determination area setting means set as the color determination area and the area set by the color determination area setting means are read, and the distribution center of the background color and the ruled line color in the RGB space is connected by a line from the area to determine the ruled line color. A color information storage unit 2S for storing the obtained information and a ruled line color reference line determining means for forming a reference line, and a character area including the ruled line, and its distribution and the distance between the ruled line color reference line. And a comparison means for comparing whether or not the distance calculated by the distance calculation section is larger than a predetermined threshold value, and when the distance is larger, it is extracted as character information. did A character extraction unit 2, a binarization unit 3 binarizes the character information extracted by the character extracting unit, binarization processing section recognizes the binary information formed by three character recognition unit 4
And characterized in that:
【００１４】この装置は図２に示すように罫線を含む原
画像から、ＲＧＢ空間での背景色と罫線色の分布中心か
らの距離値を求め、距離値がしきい値よりも大きいとき
のみその情報を取り出すようにしこれを文字情報として
二値化して取り出すものである。As shown in FIG. 2, this apparatus obtains the distance value from the distribution center of the background color and the ruled line color in the RGB space from the original image including the ruled line, and only when the distance value is larger than the threshold value, Information is extracted and binarized as character information.
【００１５】この装置を用いて、まず、図３に示すよう
な伝票から文字情報を抽出する場合について説明する。First, a case where character information is extracted from a slip as shown in FIG. 3 using this apparatus will be described.
【００１６】まずここでは“納品書”の“納”（図４）
という文字を含むエリアを色情報検出エリア（色判定領
域）とし、認識対象の文字領域（文字認識エリア）とし
て、日付エリアを読み取る場合を考える。なおこの
“納”という文字の色は日付けエリア内の罫線の色と同
色である。図５は、全体の処理を示すフローチャートで
ある。ここでは、画像入力を行い（ステップ１０１）、
これらの入力情報から文字抽出を行い（ステップ１０
２）２値化して（ステップ１０３）、文字認識を行う
（ステップ１０４）。First, here, "delivery" of the "delivery slip" (Fig. 4)
Consider a case in which the area including the character is a color information detection area (color determination area) and the date area is read as a character area to be recognized (character recognition area). Note that the color of the character "NA" is the same as the color of the ruled line in the date area. FIG. 5 is a flowchart showing the overall processing. Here, an image is input (step 101),
Characters are extracted from these input information (step 10
2) Binarization (step 103) and character recognition (step 104).
【００１７】ここで文字抽出ステップにおいては、まず
あらかじめ定められた色情報検出エリアから色情報を検
出し、ＲＧＢそれぞれの色について罫線色と背景色とに
ついてヒストグラム（図６）を作成し、これからＲＧＢ
空間分布図を形成する。そして認識対象エリアの読取り
を行い、文字情報についてのＲＧＢ空間分布図を形成す
る（図７）。Here, in the character extraction step, first, color information is detected from a predetermined color information detection area, a histogram (FIG. 6) is created for ruled line colors and background colors for each of the colors RGB, and RGB is then created.
Form a spatial distribution map. Then, the recognition target area is read to form an RGB space distribution map for character information (FIG. 7).
【００１８】そしてこのＲＧＢ空間分布図に基づいて、
罫線色基準線からの距離の分布を算出し、距離分布図を
作成する（図８）。この距離分布図からしきい値を決定
し、このしきい値よりも小さいときは罫線情報であると
し、しきい値よりも小さい情報を除いたものを文字情報
として取り出す。これにより図９(a) に示すように罫線
および文字を含んだ原画像から、罫線情報を除いて図９
(b) に示すような２値化画像を取り出すことができる。Then, based on this RGB space distribution map,
The distribution of the distance from the ruled line color reference line is calculated to create a distance distribution map (FIG. 8). A threshold value is determined from this distance distribution map, and when it is smaller than this threshold value, it is regarded as ruled line information, and information smaller than the threshold value is removed as character information. This removes ruled line information from the original image containing ruled lines and characters as shown in FIG.
A binarized image as shown in (b) can be taken out.
【００１９】次に図１０にもとづいて文字抽出について
詳細に説明する。前述したように画像入力ステップ１０
１で入力された画像から色情報検出エリアの切り出しを
行う（ステップ２０１）。ここでは納品書の“納”とい
う文字を選択して色情報検出エリアとして切り出す。そ
してこの色情報検出エリアの入力画像についてＲＧＢそ
れぞれの濃度ヒストグラムを作成し（ステップ２０
２）、ヒストグラムの極大点を検出する（ステップ２０
３）。なお、この背景−罫線の直線検出に際してのＲＧ
Ｂそれぞれの色について罫線色と背景色とについて検出
したヒストグラムは図１１に示すごとくである。ここで
ｒ１，ｇ１，ｂ１はそれぞれ罫線色を示し、ｒ２，ｇ
２，ｂ２はそれぞれ背景色を示す。上述したように
“納”という文字の色が文字認識エリア内の罫線の色に
等しいので、ここでは“納”の文字色を罫線色と表現し
ている。なお濃度は例えば、ｒ１＞ｒ２，ｇ１＞ｇ２，
ｂ１＞ｂ２である。Next, the character extraction will be described in detail with reference to FIG. As described above, the image input step 10
The color information detection area is cut out from the image input in step 1 (step 201). Here, the character "delivery" on the delivery note is selected and cut out as a color information detection area. Then, a density histogram for each of RGB is created for the input image in the color information detection area (step 20).
2) The maximum point of the histogram is detected (step 20).
3). In addition, this background-RG at the time of detecting the straight line of the ruled line
The histograms detected for the ruled line color and the background color for each of the colors B are as shown in FIG. Here, r1, g1, and b1 respectively indicate ruled line colors, and r2, g
2 and b2 respectively indicate background colors. As described above, the color of the character "NAO" is equal to the color of the ruled line in the character recognition area, so the character color of "NAO" is expressed as the ruled line color here. The concentration is, for example, r1> r2, g1> g2.
b1> b2.
【００２０】このようにして背景−罫線の中心座標のＲ
ＧＢ情報（ｒ１，ｇ１，ｂ１，ｒ２ｇ２，ｂ２）を得る
（ステップ２０４）。In this way, R of the center coordinate of the background-ruled line is
GB information (r1, g1, b1, r2g2, b2) is obtained (step 204).
【００２１】次に、文字認識エリアここではたとえば日
付エリアを選択して文字認識エリアとして切り出す（ス
テップ２０５）。Next, a character recognition area, for example, a date area is selected and cut out as a character recognition area (step 205).
【００２２】そして、罫線情報と文字情報との混在した
領域である文字認識エリアを読み取った場合の出力を各
色についてのヒストグラムを３次元で表したものを図１
２に示す。ここでは背景色の点Ｏを頂点とするＶ字状を
示す。Ｋは罫線情報、Ｃは文字情報に相当する。このよ
うに、罫線情報と文字情報とははっきりとわかれるた
め、あるしきい値をあらかじめ設定しておけば、容易に
抽出可能である。なお、図中複数の点が重なるものも１
つの点として表示しており、必ずしも密度は表現されて
いない。FIG. 1 shows a three-dimensional histogram of the output for each color when the character recognition area, which is a mixed area of ruled line information and character information, is read.
It is shown in FIG. Here, a V-shape having a vertex O of the background color is shown. K corresponds to ruled line information, and C corresponds to character information. In this way, since the ruled line information and the character information are clearly distinguished, it is possible to easily extract them by setting a certain threshold value in advance. It should be noted that the number of overlapping points in the figure is 1
It is shown as one dot, and the density is not always expressed.
【００２３】この図１２における分布図から、距離演算
部２Ｃにおいて前記罫線色基準線との距離を演算（座標
変換）する（ステップ２０６）（図１３）。罫線Ａ（ｒ
１，ｇ１，ｂ１） 背景Ｂ（ｒ２，ｇ２，ｂ２） 文字
Ｃとしたとき、文字Ｃと罫線色基準線との距離ｈは次式
で表される。From the distribution chart in FIG. 12, the distance calculation unit 2C calculates (coordinates) the distance from the ruled line color reference line (step 206) (FIG. 13). Ruled line A (r
1, g1, b1) Background B (r2, g2, b2) When the character is C, the distance h between the character C and the ruled line color reference line is expressed by the following equation.
【００２４】 このようにして抽出した文字情報に対して距離値ヒスト
グラムを作成（ステップ２０７）する（図８）。[0024] A distance value histogram is created for the character information thus extracted (step 207) (FIG. 8).
【００２５】そしてあとは前述したように二値化処理
（ステップ１０３）を経て文字認識を行う（ステップ１
０４）。After that, character recognition is performed through the binarization process (step 103) as described above (step 1).
04).
【００２６】このようにして多色情報から文字情報のみ
を高速で容易に抽出して認識することが可能となる。In this way, it becomes possible to easily extract and recognize only the character information from the multicolor information at high speed.
【００２７】[0027]
【発明の効果】以上説明してきたように、本発明によれ
ば、多色印刷においても極めて容易に高速でかつ安定し
て文字抽出を行うことが可能となる。As described above, according to the present invention, even in multicolor printing, it is possible to extract characters very quickly and stably.
【図１】本発明の装置の機能ブロック図FIG. 1 is a functional block diagram of an apparatus of the present invention.
【図２】本発明による抽出過程を示す図FIG. 2 is a diagram showing an extraction process according to the present invention.
【図３】読み取るべき伝票を示す図FIG. 3 is a diagram showing a slip to be read.
【図４】背景・罫線色判定エリアを示す図FIG. 4 is a diagram showing a background / ruled line color determination area
【図５】本発明による文字抽出工程のフローチャート図FIG. 5 is a flowchart of a character extraction process according to the present invention.
【図６】Ｒのヒストグラムを示す図FIG. 6 is a diagram showing a histogram of R.
【図７】ＲＧＢ空間分布を示す説明図FIG. 7 is an explanatory diagram showing an RGB space distribution.
【図８】距離分布を示す図FIG. 8 is a diagram showing a distance distribution
【図９】原画像と処理後の二値化画像を示す図FIG. 9 is a diagram showing an original image and a binarized image after processing.
【図１０】文字抽出のフローチャートを示す図FIG. 10 is a diagram showing a flowchart of character extraction.
【図１１】背景−罫線基準線検出のためのＲＧＢそれぞ
れの色について罫線色と背景色とについて検出したヒス
トグラムを示す図FIG. 11 is a diagram showing a histogram detected for a ruled line color and a background color for each of RGB colors for background-ruled line reference line detection.
【図１２】認識対象エリアの入力画像をＲＧＢ空間分布
として示す図FIG. 12 is a diagram showing an input image of a recognition target area as an RGB spatial distribution.
【図１３】背景−罫線基準線からの距離の算出過程を示
す図FIG. 13 is a diagram showing a process of calculating a distance from a background-ruled line reference line.
１ 画像入力部 ２ 文字抽出部 ３ 二値化処理部 ４ 文字認識部 1 image input unit 2 character extraction unit 3 binarization processing unit 4 character recognition unit
Claims (1)
報のみを取り出す文字抽出装置において、 背景色と罫線枠の色のみが存在する領域を色判定領域と
して設定する色判定領域設定手段と、 前記色判定領域設定手段で設定された領域を読取り、そ
の領域からＲＧＢ空間での背景色と罫線色の分布中心を
線で結び、罫線色を示す基準線を形成する罫線色基準線
決定手段と、 罫線を含む文字領域を読取り、その分布と前記罫線色基
準線との距離を演算する演算手段と、 前記演算手段によって算出された距離があらかじめ決定
されたしきい値よりも大きいか否かを比較する比較手段
とを含み、大きいときのみ文字情報としてとりだすよう
にしたことを特徴とする文字抽出装置。1. A character extraction device for extracting only character information from a document such as a slip having a ruled line frame, and a color determination region setting means for setting a region in which only the background color and the color of the ruled line frame are present as a color determination region, A ruled line color reference line determination unit that reads the region set by the color determination region setting unit, connects the background color in RGB space and the distribution center of the ruled line color with a line from the region, and forms a reference line indicating the ruled line color; , A character area including a ruled line, calculating means for calculating the distance between the character area including the ruled line and the ruled line color reference line, and determining whether the distance calculated by the calculating means is larger than a predetermined threshold value. A character extracting device including a comparing means for comparing and extracting as character information only when it is large.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP7251395A JPH0997309A (en) | 1995-09-28 | 1995-09-28 | Character extracting device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP7251395A JPH0997309A (en) | 1995-09-28 | 1995-09-28 | Character extracting device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| JPH0997309A true JPH0997309A (en) | 1997-04-08 |
Family
ID=17222212
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| JP7251395A Pending JPH0997309A (en) | 1995-09-28 | 1995-09-28 | Character extracting device |
Country Status (1)
| Country | Link |
|---|---|
| JP (1) | JPH0997309A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2008310817A (en) * | 2007-06-18 | 2008-12-25 | Sharp Corp | Method for detecting line structure from text map and image processor |
| JP2009159185A (en) * | 2007-12-26 | 2009-07-16 | Hitachi Computer Peripherals Co Ltd | Drop out color treatment method and its apparatus |
| US7729534B2 (en) | 2004-03-22 | 2010-06-01 | Canon Kabushiki Kaisha | Image-processing device and image-processing method for extracting a recognition-target area including a character from a target image |
| JP2011018359A (en) * | 2010-09-16 | 2011-01-27 | Hitachi Computer Peripherals Co Ltd | Dropout color processing apparatus and dropout color processing method using the same |
| JP2012095002A (en) * | 2010-10-25 | 2012-05-17 | Hitachi Omron Terminal Solutions Corp | Image correction device and image correction method |
| US9092892B2 (en) | 2012-03-30 | 2015-07-28 | Brother Kogyo Kabushiki Kaisha | Image processing device determining binarizing threshold value |
| US9230186B2 (en) | 2012-03-30 | 2016-01-05 | Brother Kogyo Kabushiki Kaisha | Image processing device generating binary image data by selecting specific component |
-
1995
- 1995-09-28 JP JP7251395A patent/JPH0997309A/en active Pending
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7729534B2 (en) | 2004-03-22 | 2010-06-01 | Canon Kabushiki Kaisha | Image-processing device and image-processing method for extracting a recognition-target area including a character from a target image |
| JP2008310817A (en) * | 2007-06-18 | 2008-12-25 | Sharp Corp | Method for detecting line structure from text map and image processor |
| US8000535B2 (en) | 2007-06-18 | 2011-08-16 | Sharp Laboratories Of America, Inc. | Methods and systems for refining text segmentation results |
| JP2009159185A (en) * | 2007-12-26 | 2009-07-16 | Hitachi Computer Peripherals Co Ltd | Drop out color treatment method and its apparatus |
| JP4654237B2 (en) * | 2007-12-26 | 2011-03-16 | 日立コンピュータ機器株式会社 | Dropout color processing device |
| JP2011018359A (en) * | 2010-09-16 | 2011-01-27 | Hitachi Computer Peripherals Co Ltd | Dropout color processing apparatus and dropout color processing method using the same |
| JP2012095002A (en) * | 2010-10-25 | 2012-05-17 | Hitachi Omron Terminal Solutions Corp | Image correction device and image correction method |
| US9092892B2 (en) | 2012-03-30 | 2015-07-28 | Brother Kogyo Kabushiki Kaisha | Image processing device determining binarizing threshold value |
| US9230186B2 (en) | 2012-03-30 | 2016-01-05 | Brother Kogyo Kabushiki Kaisha | Image processing device generating binary image data by selecting specific component |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104899586B (en) | Method and device is identified to the word content for including in image | |
| EP1310912A2 (en) | Image processing method, apparatus and system | |
| JP3018949B2 (en) | Character reading apparatus and method | |
| JPH0997309A (en) | Character extracting device | |
| EP1445100A2 (en) | Printed paper inspecting method and apparatus | |
| JP2003087562A (en) | Image processor and image processing method | |
| JP3955467B2 (en) | Image processing program and image processing apparatus | |
| JP3276555B2 (en) | Format recognition device and character reader | |
| JP2008252239A (en) | Document processor | |
| JP2005071134A (en) | Optical character reader | |
| JP3756660B2 (en) | Image recognition method, apparatus and recording medium | |
| JP2001195584A (en) | Image recognizing device | |
| JP3412441B2 (en) | Image processing device | |
| JPH05128308A (en) | Character recognition device | |
| JPH08101880A (en) | Character recognition device | |
| JPH0749874A (en) | Method for preparing index | |
| JP2001216473A (en) | Method and device for processing image and recording medium | |
| JP2000331118A (en) | Image processor and recording medium | |
| JP2003123076A (en) | Image processor and image processing program | |
| JPS6361387A (en) | Character segmenting system | |
| JPH06266887A (en) | Method for preparing format and optical character reader using the method | |
| JPS58222381A (en) | Character stubbing system | |
| JP2004240500A (en) | Device and program for processing image, and storing medium | |
| JPH05290212A (en) | Character recognition device | |
| JPH06215139A (en) | Graphic recognizing method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| A977 | Report on retrieval |
Free format text: JAPANESE INTERMEDIATE CODE: A971007 Effective date: 20040831 |
|
| A131 | Notification of reasons for refusal |
Free format text: JAPANESE INTERMEDIATE CODE: A131 Effective date: 20041102 |
|
| A521 | Request for written amendment filed |
Free format text: JAPANESE INTERMEDIATE CODE: A523 Effective date: 20041224 |
|
| A02 | Decision of refusal |
Free format text: JAPANESE INTERMEDIATE CODE: A02 Effective date: 20050222 |