JPH09167206A - Space detecting method for japanese/english-mixed document, pitch format judging method, space detecting method for constant pitch alphanumeric character string and space detecting method for proportional pitch alphanumeric character string - Google Patents

Abstract

PROBLEM TO BE SOLVED: To detect space in an alphanumeric character string which exits mixedly with a Japanese character string with high precision by picking-up the alphanumeric character string from a character recognition result as an object range, discriminating a pitch format at every object area and executing a space detection processing by pitch format for every object range. SOLUTION: At first, in an object range pick-up part 204, the character recognition result inside a result memory 211 is referred and the alphanumeric string as the object range of the space detection processing is picked-up. Information concerning the picked-up object range is preserved in a work memory 212. Then, in a pitch format judging part 205, the contents of a segmenting memory 209 and the result memory 211 are referred concerning the character of the object range and it is judged whether the object range is a proportional pitch or a constant pitch. The judgement result is preserved in the work memory 212. A constant pitch character string space detection processing or a proportional pitch character string space processing is executed in accordance with the judgement result.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a space detection process in a character recognition device, and more particularly to a space detection process for a Japanese / English mixed document.

[0002]

2. Description of the Related Art When converting a document into text data by character recognition so that the document can be reproduced from the text data, it is necessary to detect spaces in the document during character recognition. The following is known as a conventional technique relating to such space detection in a document.

Prior art 1: Counting the number of white pixels between characters,
The number of space codes is determined by dividing the count value by the standard character pitch (Japanese Patent Application No. 63-14182).

Prior art 2: Determine the pitch format (constant pitch / proportional pitch) of a document. Then, the space is detected by a method according to the determined pitch format (Japanese Patent Application No. 1-161176).

Prior art 3: In a Japanese sentence in which Japanese and English characters are mixed, the space width between characters and the character width information are compared to determine the type of space and the presence / absence of the space, but the types of characters before and after the space are referred to. To determine whether the space is valid or invalid (Japanese Patent Application No. 2-214136).

Prior art 4: In a Japanese sentence containing both Japanese and English characters, the space width between characters and the standard size of the character are compared to determine the type and presence of the space. Depending on the character adjacent to the space between characters, The space width between characters is corrected, and the corrected value is compared with the standard character size (Japanese Patent Application No. 3-18476).
issue).

[0007]

[Problems to be Solved by the Invention] In a document in which Japanese characters and alphanumeric characters (English characters and numbers) are mixed, the difference between the fonts of Japanese characters and alphanumeric characters, the difference in character size between full-width characters and half-width characters, and constant pitch There is a difference in pitch format called proportional pitch, and such a difference occurs in an unspecified part of the document. And the width of the space also differs depending on the location. This can be easily understood by looking at the print sample shown in FIG. In FIG. 9, in Sample 1 in which all double-byte characters are printed at a constant pitch, “R
There is a space only between the "h" of "icoh" and the next "P". On the other hand, in Sample 2 in which alphanumeric characters were printed at a proportional pitch, spaces were placed between the "new model" and "R", and between "Ricoh" and "h" and "P". And the space is narrower than the space of Sample 1.

For such a Japanese-English mixed document, it has been difficult in the prior art to detect the space with high accuracy.
Further, there is no known method for accurately determining the pitch format when Japanese characters and alphanumeric characters are mixed in a line. Prior art 2 assumes that the entire document or the entire line is an English sentence and processes it in units of English words. Therefore, it is not possible to accurately detect a space in an English word that partially appears in a line. Regarding space detection in constant pitch character string,
Since the character spacing of the constant-pitch character string fluctuates greatly, even if full-width spaces can be detected, the conventional art 4 for correcting the character spacing also fails to detect half-width spaces. Regarding the space detection in the proportional pitch character string, since the character spacing of the proportional pitch character string is narrow, the space cannot be detected by the prior art 4.

The object of the present invention is suitable for an improved method capable of detecting a space with high precision in a Japanese-English mixed document in which Japanese characters and alphanumeric characters are mixed in a line, and the space detecting method. Another object of the present invention is to provide a method for determining a proper pitch format, a method for detecting a space in a constant pitch character string, and a method for detecting a space in a proportional pitch character string.

[0010]

According to the space detection method for a Japanese-English mixed document according to the invention described in claim 1, a character recognition device for recognizing a character from an image of a document recognizes the character recognition result of the English-Japanese mixed document. A target range extraction process for extracting an alphanumeric character string based on the target range, and a pitch format determination process for determining whether the pitch format is a constant pitch or proportional pitch for each target range extracted by the target range extraction process, Constant pitch character string space detection processing for detecting a space for each target range determined to be a constant pitch by the pitch format determination processing, and space detection for each target range determined to be a proportional pitch by the pitch format determination processing Proportional pitch character string space detection processing is performed.

According to a second aspect of the present invention, in the space detection method for a Japanese-English mixed document according to the first aspect of the present invention, the pitch format determination processing determines each target range based on cutout information for each target range. If the character rectangle interval ratio, which is the ratio of the character rectangle interval to the standard character size, is smaller than a predetermined threshold value, if the character rectangle interval ratio is larger than the character rectangle interval of the threshold value or more, the proportional pitch is The feature is that it is determined, and if it is not, the pitch is determined to be constant.

According to a third aspect of the present invention, in the space detection method for a Japanese-English mixed document according to the first aspect, the constant pitch character string space detection process is performed to detect a character rectangular pitch between characters before and after a focused character. There is a space between the focused characters when the smaller character rectangle pitch among the focused characters is used as the reference pitch between the focused characters and the ratio of the rectangular character pitch between the focused characters to the reference pitch is greater than a predetermined threshold. Then, the determination is made.

According to a fourth aspect of the present invention, in the space detection method for a Japanese-English mixed document according to the first aspect of the invention, the standard is determined in each proportional range based on the cutout information in the proportional pitch character string space detection processing. When the character rectangle interval ratio, which is the ratio of the character rectangle interval to the character size, is larger than a predetermined threshold value, it is determined that there is a space between the corresponding characters.

According to a fifth aspect of the pitch format determining method of the present invention, in a character recognizing device for recognizing a character cut out from a document image, for each recognized alphanumeric character string, cutout information is obtained for each alphanumeric character string. If the character rectangle interval ratio, which is the ratio of the character rectangle interval to the standard character size determined based on, is smaller than a predetermined threshold value, the character rectangle interval ratio is larger than the character rectangle interval of the threshold value or more. For example, it is determined to be proportional pitch, and if not, it is determined to be constant pitch.

According to a sixth aspect of the present invention, there is provided a space detecting method for a constant-pitch alphanumeric character string, wherein a character recognizing device for slicing and recognizing characters from an image of a document recognizes each constant-pitch alphanumeric character string. The smaller character rectangle pitch of the character rectangle pitch between the characters before and after the selected character is set as the reference pitch between the focused characters, and the ratio of the character rectangular pitch between the focused characters to the reference pitch is less than the predetermined threshold. When it is large, it is characterized in that it is determined that there is a space between the noted characters.

According to a seventh aspect of the present invention, there is provided a space detecting method for a proportional pitch alphanumeric character string in a character recognition device for recognizing a character cut out from an image of a document, in which each alphanumeric character string of the proportional pitch is recognized.
When the character rectangle interval ratio, which is the ratio of the character rectangle interval to the standard character size determined based on the cutout information for each alphanumeric character string, is larger than a predetermined threshold value, it is determined that there is a space between the corresponding characters. It is characterized by that.

[0017]

BEST MODE FOR CARRYING OUT THE INVENTION An embodiment of the present invention will be described with reference to the drawings in order to clarify the embodiment of the present invention. FIG. 1 shows an overall processing flow of an embodiment of the present invention, and FIG.
Shows an example of a device configuration for one embodiment of the present invention. The flow of the space detection process in FIG. 1 is shown in FIG. The flow of the pitch format determination process in FIG. 3 is shown in FIG. 4, the flow of the proportional pitch character string space detection process is shown in FIG. 5, and the flow of the constant pitch character string space detection process is shown in FIG. 7 and 8 are diagrams for explaining the space detection process.

First, referring to FIGS. 1 and 2, the overall processing flow will be described in association with the apparatus configuration. First, data of a document image is input by the image input unit 200 and stored in the image memory 208 via the bus 213 (step 100). When an image is input, the line / character cutout unit 202 cuts out character lines and characters in the document image in the image memory 208, and stores information on the cut-out character lines and character areas in the cutout memory 209 (step 110). The cutout information stored here is, for example, the coordinates of the start and end points of a character line or the start point coordinates and width,
It is the coordinates and size information of the diagonal vertices of the circumscribing rectangle (character rectangle) of the character. It goes without saying that these pieces of information are created and stored in such a form that the correspondence between each character rectangle and the character line to which it belongs can be identified. Then, the character recognition unit 203 refers to the cut-out information in the cut-out memory 209, reads the character image from the image memory 208, and compares it with the character dictionary in the character dictionary memory 210 to determine the character code for the character image. Then, it is stored in the result memory 211 (step 120).
The character recognition unit 203 may use language knowledge such as words and grammars to perform post-processing such as correction of a recognition result by comparison with a character dictionary.

When the recognition processing by the character recognition unit 203 is completed for the whole or a part of the document image, the space detection processing (step 130) is started. This space detection processing (step 130) is performed by the target range extraction unit 20.
4. The pitch format determination unit 205, the constant pitch character string space detection unit 206, and the proportional pitch character string space detection unit 207 are executed, and the cutout memory 209, the result memory 211, and the work memory 212 are referred to at that time. The space detection result is stored in the result memory 211. When the space detection process ends, the data in the result memory 211 is output to the outside (step 14
0), the entire process ends.

The outline of the space detection process (step 130) is as shown in FIG. First, the target range extraction unit 2
In 04, the character recognition result (character code) in the result memory 211 is referred to, and the alphanumeric character string as the target range of the space detection processing is extracted (step 301). Information about the extracted target range is stored in the work memory 212. Next, the pitch format determination unit 205 refers to the contents of the cutout memory 209 and the result memory 211 for the characters in the target range, and determines whether the target range is the proportional pitch or the constant pitch (step 301). The determination result is stored in the work memory 212. For the target range determined as the proportional pitch, the proportional pitch character string space detection unit 207 cuts out the memory 209 and the result memory 2.
Referring to the contents of No. 11, the space is detected by the method for proportional pitch, and the detected space is written in the result memory 211 as a space code (step 302). For the target range determined to be the constant pitch, the constant pitch character string space detection unit 206 refers to the contents of the cutout memory 209 and the result memory 211 to detect the space by the method for the constant pitch and detect the space. Result memory 211 with space as space code
(Step 303). Specific contents of the processing described above will be described with reference to FIGS. 4 to 8.

Consider the case of processing a character line as shown in FIG. This character line has two parts including spaces. One is “Preter 5” printed with proportional pitch (Times-Roman font).
It is between “Preter” of “5” and “55”. The other is "Imagio" and "77" of "Imagio 77" printed at a constant pitch (the font is Courier).
Between. Character recognition for this character line (Step 1
The result of 20) does not include spaces, and says "new copy, Pr
eter55 and Image77 released ”.

In the target range extraction processing (step 300), a character string in which five or more alphanumeric characters (alphabetic characters or numbers) are consecutive is extracted as a target range by referring to the character recognition result that does not include such spaces. . Therefore, in the character line shown in FIG. 7A, as shown in FIG. 7B, two character strings of “Preter 55” and “Imagio 77” are extracted as the target range. The character length for extraction determination is not limited to 5 characters, and may be appropriately changed according to the type of document to be processed. Further, the method of determining the target range itself can be changed appropriately.

With respect to the alphanumeric character string thus extracted as the target range, pitch format judgment (step 30
Perform 1). FIG. 4 is a flowchart thereof.

First, the standard character size for the target range is determined (step 400). Specifically, 1.25 times the maximum character rectangle height in the target range (character size A) and the maximum character rectangle height in the character line that includes the target range (character size B) are obtained. The larger value of the character sizes A and B is the standard character size. However, if there is a character rectangle of "j" in the target range, the character rectangle is not used for determining the standard character size. FIG. 8 is an explanatory diagram of such standard character size determination.

In the first target range shown in FIG. 7,
Since the height of the "P" character rectangle is the maximum, compare that value multiplied by 1.25 with the maximum "new" character rectangle height in the character line, and the larger value is the standard. It is assumed to be the font size. The standard character size may be determined by another method.

After determining the standard character size for the target range as described above, next, using this standard character size, a pair of adjacent character rectangles is selected from the beginning of the target range, and a space between the character rectangles is selected. The width of the part, that is, the ratio of the character rectangle interval (see FIG. 7C) to the standard character size (character rectangle interval ratio = character rectangle interval / standard character size) is calculated (steps 405 and 410). The calculated character rectangle interval ratio is stored in the work memory 212. The character rectangle interval is calculated from the coordinates of the two character rectangles stored in the cutout memory 209. However, the character rectangle interval may be obtained in the cutting process stage, and in that case, the character rectangle interval read from the cutout memory 209 may be used for the calculation of the character rectangle interval ratio. Next, the character rectangle interval ratio calculated in this way is set to a predetermined threshold value TH1.
(Step 415). This threshold TH1 is, for example, 0.17 (not limited to this value, it was confirmed by experiments that a value around this is appropriate in general documents. However, it should be adjusted according to the document to be processed. Is preferred). When the character rectangle interval ratio is smaller than the threshold value TH1, the counter NS (initial value is 0) is incremented by 1 (step 420), and when the character rectangle interval ratio is equal to or greater than the threshold value TH1, the counter NL (initial value is 0) is increased by 1. Incremented only (step 421). If the next character rectangle pair remains, steps 405 to 420
The processing loop of is repeated.

When the processing is completed up to the last character rectangle pair, the processing loop from step 405 to step 420 or 421 is exited and the process proceeds to step 425. In this step 425, the value of the counter NS and the value of the counter NL are compared (step 425). The value of the counter NS is the number of character rectangle intervals in which the ratio of character rectangle intervals in the target range is smaller than the threshold value TH1, that is, the number of character rectangle intervals narrower than “standard character size * TH1”, and the value of the counter NL is the number of wide character rectangle intervals. . Therefore, in step 425, it is checked whether the narrow character rectangle interval is a majority. NS
> NL, that is, when it is determined that the narrow character rectangle interval is a majority, the pitch format of the character string in the target range is set to proportional pitch, and the flag information is set to the work memory 212.
Write in association with the target range (step 430),
If not, that is, if the wide character rectangle interval is less than half, the pitch format is set to constant pitch, and the flag information is written in the work memory 212 in association with the target range (step 435). According to this flag information, the proportional pitch character string space detection processing (step 302) or the constant pitch character string space detection processing (step 303) is selected as the next processing.

In the example shown in FIG. 7A, the first target range "Preter55" is determined to be a proportional pitch as shown in FIG. 7C, and the other target range "Imagio77" is a constant pitch. To be judged.

The processing contents of the space detection processing (step 302) for the target range determined as the proportional pitch will be described. FIG. 5 is a flowchart thereof.

The character rectangle interval ratio of the target range stored in the work memory 212 is taken out and is taken as a predetermined threshold value T.
It is compared with H2 (steps 500 and 505). The threshold TH2 is, for example, 0.29 (not limited to this value,
In general documents, it was confirmed by experiments that the value around this is appropriate. However, it is preferable to adjust according to the document to be processed). If the character rectangle interval ratio is TH2 or less, that is, the character rectangle interval is “standard character size * TH
When it is narrower than 2 ”, it is determined that there is no space between the corresponding character rectangle pairs. When the character rectangle interval ratio is larger than TH2, that is, when the character rectangle interval is wider than “standard character size * TH2”, it is determined that there is a space between the corresponding character rectangle pairs, and between the corresponding recognition result character codes. A space code is inserted (step 510). Similar processing is executed for all character rectangle interval ratios within the target range.

In the example shown in FIG. 7 (a), as shown in FIG. 7 (d), the "r" at the end of "Preter" in the target range determined to be proportional pitch.
Since only the character rectangle interval ratio between the next and "5" exceeds TH2, the space code is inserted at that position.

The processing contents of the space detection processing (step 303) for the target range determined to be the constant pitch will be described. FIG. 6 is a flowchart thereof.

The character rectangle pitch between all the characters within the target range is calculated, and the result is stored in the work memory 212 (step 600). That is, the centers of adjacent character rectangles between characters are calculated, and the distance between the centers is determined as the character rectangle pitch between the characters. For this calculation, the coordinate data of the character rectangle stored in the cutout memory 209 is used. However, the center interval of the adjacent character rectangles may be calculated in advance in the cutout processing stage, stored in the cutout memory 209, and the value may be read out.

When the calculation of the character rectangular pitch is completed, one character of interest is selected from the beginning of the target range (step 605), and the character rectangular pitch between the characters before the character of interest and the character of interest are selected. The character rectangular pitch between the characters after the interval is compared, and the smaller character rectangular pitch is used as the reference pitch for the target character (step 610). When the target character is the first character in the target range, there is no character space before the target character, and therefore the character rectangular pitch between the characters after the target character is set as the reference pitch. Similarly,
When the space between the characters of interest is between the last characters in the target range,
The character rectangular pitch between the preceding characters is used as the reference pitch.

Next, a value obtained by dividing the character rectangle pitch between the target characters by the reference pitch is calculated as a character rectangle pitch ratio (= character rectangle pitch / reference pitch) (step 615).
Then, it is determined whether the character rectangle pitch ratio is larger than a predetermined threshold value TH3, that is, whether the character rectangle pitch between the target characters is wider than "reference pitch * TH3" (step 62).
0). The threshold value TH3 is, for example, 1.8 (not limited to this value, but in general documents, a value around this value can detect both a full-width space in a full-width character string and a half-width space in a half-width character string, and is appropriate. However, it is preferable to adjust it according to the document to be processed).

If the character rectangle pitch ratio is TH3 or less,
It is determined that there is no space between the characters of interest. When the character rectangle pitch ratio is greater than TH3, it is determined that there is a space between the characters of interest, so the result memory 2
A space code is inserted between the corresponding character codes in 11 (step 625). The same process is repeated until the last character in the target range.

In the example shown in FIG. 7A, as shown in FIG. 7D, the last "o" of "Imagio" and the next "7" in the target range determined to have a constant pitch. Since the character rectangle pitch ratio exceeds TH3 only between the characters, it is determined that there is a space between the characters, and the space code is inserted.

[0038]

According to the first aspect of the invention, an alphanumeric character string is extracted as a target range from the character recognition result, the pitch format is determined for each target range, and the pitch format is determined for each target range. Separate space detection processing is performed, so a mixture of alphanumeric characters and Japanese characters in a line, and a mixture of fixed-pitch alphanumeric characters and proportional-pitch alphanumeric characters in a Japanese-English document Even with respect to, the space can be detected with high accuracy.

According to the second aspect of the invention, the standard character size is determined for each alphanumeric character string extracted as the target range,
Since the pitch format is determined by paying attention to the relative ratio of the character rectangle spacing in the alphanumeric character string to the standard character size, the alphanumeric character string is mixed with the Japanese character string, and the full-width character and the half-width character Even when several character strings are mixed, it is possible to accurately determine whether the pitch format of the alphanumeric character string is constant pitch or proportional pitch. Therefore, for alphanumeric character strings in a Japanese-English mixed document, Since the space detection processing that matches the pitch format can be applied accurately,
It is possible to detect with high precision a space in a document in which a constant pitch and a proportional pitch alphanumeric character string are mixed.

Since the character rectangular pitch of the constant-pitch alphanumeric character string varies greatly depending on the character, even if the relative width between the constant reference pitch and the character rectangular pitch that does not reflect this variation is checked, the space is detected accurately. Can not. On the other hand, according to the third aspect of the invention, the smaller one of the character rectangular pitches between the characters before and after the focused character is used as the reference pitch for the focused character. It is possible to accurately detect spaces in a constant pitch alphanumeric character string with a large variation in, and thus to improve space detection accuracy for a document in which a constant pitch alphanumeric character string is mixed.

Since the character-pitch alphanumeric character string has a narrow character-rectangle interval, the method of comparing the standard character size and the character-rectangle interval without considering the pitch format and the character type is as follows.
Space detection was difficult. According to the invention described in claim 4, the standard character size is determined for each individual proportional pitch alphanumeric character string, and the ratio of the character rectangle interval to the standard character size is thresholded. The space can be detected with high accuracy, and therefore the space detection accuracy of a document in which an alphanumeric character string of proportional pitch is mixed can be improved.

According to the fifth aspect of the present invention, the standard character size is determined for each alphanumeric character string, and the pitch is focused on the relative wideness of the character rectangle spacing in the alphanumeric character string with respect to the standard character size. In order to determine the format, even if the alphanumeric character string is mixed with the Japanese character string, and also if the full-width character and half-width character alphanumeric character string are mixed, the alphanumeric character string has a constant pitch or proportional pitch. Can be accurately determined.

Since the character rectangular pitch of the constant-pitch alphanumeric character string varies greatly depending on the character, even if the relative width between the constant reference pitch and the character rectangular pitch that does not reflect this variation is examined, the space can be detected accurately. However, according to the invention of claim 6, the smaller one of the character rectangular pitches between the characters before and after the focused character is used as the reference pitch for the focused characters. It is possible to accurately detect a space in a constant pitch alphanumeric character string that greatly varies.

The proportional pitch alphanumeric character string has a narrow character rectangle interval. Therefore, in the method of comparing the standard character size and the character rectangle interval without considering the pitch format and the character type,
Although it is difficult to detect a space, according to the invention of claim 7, the standard character size is determined for each proportional pitch alphanumeric character string, and the ratio of the character rectangular interval to the standard character size is thresholded. The space in the alphanumeric string of pitch can be detected accurately.

[Brief description of the drawings]

FIG. 1 is a flowchart showing an overall processing flow of an embodiment of the present invention.

FIG. 2 is a block diagram showing an example of a device configuration for one embodiment of the present invention.

FIG. 3 is a flowchart showing an outline of space detection processing in FIG.

FIG. 4 is a flowchart showing the contents of pitch format determination processing in FIG.

5 is a flowchart showing the content of proportional pitch character string space detection processing in FIG. 3. FIG.

FIG. 6 is a flowchart showing the contents of constant pitch character string space detection processing in FIG.

FIG. 7 is a diagram for explaining a specific example of space detection.

FIG. 8 is an explanatory diagram of a method of determining a standard character size for pitch format determination.

FIG. 9 is a diagram for clarifying a problem in space detection of a Japanese-English mixed document.

[Explanation of symbols]

 200 image input unit 202 line / character cutout unit 203 character recognition unit 204 target range extraction unit 205 pitch format determination unit 206 constant pitch character string space detection unit 207 proportional pitch character string space detection unit 208 image memory 209 cutout memory 210 character dictionary memory 211 Result memory 212 Work memory 213 Bus

Claims (7)

[Claims] 1. A character recognition device for recognizing a character cut out from a document image, and a target range extraction process for extracting an alphanumeric character string as a target range based on a character recognition result of a document containing both English and Japanese, and the target range. A pitch format determination process for determining whether the pitch format is a constant pitch or a proportional pitch for each target range extracted by the extraction process, and a space is detected for each target range determined to be a constant pitch by the pitch format determination process. A fixed-pitch character string space detection process and a proportional-pitch character string space detection process that detects a space for each target range determined to be a proportional pitch by the pitch format determination process Space detection method. 2. A target range extraction process for extracting an alphanumeric character string as a target range based on a character recognition result of a document containing a mixture of English and Japanese in a character recognition device for recognizing a character extracted from a document image, and the target range. A pitch format determination process for determining whether the pitch format is a constant pitch or a proportional pitch for each target range extracted by the extraction process, and a space is detected for each target range determined to be a constant pitch by the pitch format determination process. A constant pitch character string space detection process and a proportional pitch character string space detection process for detecting a space for each target range determined to be proportional pitch by the pitch format determination process,
In the pitch format determination process, for each target range, the character rectangle interval ratio, which is the ratio of the character rectangle interval to the standard character size determined based on the cut-out information for each target range, is smaller than the predetermined threshold. A space detection method for a Japanese-English mixed document, characterized in that if the character rectangle interval ratio is more than the character rectangle interval equal to or more than the threshold value, it is determined as proportional pitch, and if not, it is determined as constant pitch. 3. A target range extraction process for extracting an alphanumeric character string as a target range based on a character recognition result of an English-Japanese mixed document in a character recognition device for recognizing a character from a document image, and the target range. A pitch format determination process for determining whether the pitch format is a constant pitch or a proportional pitch for each target range extracted by the extraction process, and a space is detected for each target range determined to be a constant pitch by the pitch format determination process. A constant pitch character string space detection process and a proportional pitch character string space detection process for detecting a space for each target range determined to be proportional pitch by the pitch format determination process,
In the constant pitch character string space detection processing, the smaller character rectangular pitch of the character rectangular pitches between the characters before and after the focused character is set as the reference pitch between the focused characters, and the character rectangular pitch between the focused characters is used. A method of detecting a space in a Japanese-English mixed document, which is characterized in that when the ratio to the reference pitch of is larger than a predetermined threshold value, it is determined that there is a space between the focused characters. 4. A character recognition device for recognizing a character cut out from a document image, and a target range extraction process for extracting an alphanumeric character string as a target range based on a character recognition result of a document containing both English and Japanese, and the target range. A pitch format determination process for determining whether the pitch format is a constant pitch or a proportional pitch for each target range extracted by the extraction process, and a space is detected for each target range determined to be a constant pitch by the pitch format determination process. A constant pitch character string space detection process and a proportional pitch character string space detection process for detecting a space for each target range determined to be proportional pitch by the pitch format determination process,
In the proportional pitch character string space detection processing, when the character rectangle interval ratio, which is the ratio of the character rectangle interval to the standard character size determined based on the cutout information for each target range, is greater than a predetermined threshold value, the corresponding character A method for detecting a space in a Japanese-English mixed document characterized by determining that a space exists between them. 5. A character recognition apparatus for recognizing a character cut out from an image of a document, for each recognized alphanumeric character string, a character for a standard character size determined based on the cutout information for each alphanumeric character string. If the character rectangle interval ratio, which is the ratio of the rectangle intervals that is smaller than the predetermined threshold value, is larger than the character rectangle interval that is greater than the threshold value, the character rectangle interval ratio is determined to be proportional pitch. A pitch format determination method characterized by determining a constant pitch. 6. A character recognizing device that cuts out characters from an image of a document and recognizes, in each recognized alphanumeric character string of a constant pitch, the smaller one of the character rectangular pitches between the characters before and after the noted character. When the ratio of the character rectangular pitch between the focused characters to the reference pitch is larger than a predetermined threshold, it is determined that there is a space between the focused characters. Characteristic constant pitch alphanumeric character space detection method. 7. A character recognition device for recognizing characters from an image of a document, for each recognized alphanumeric character string of proportional pitch, with respect to a standard character size determined based on the cutout information for each alphanumeric character string. A method for detecting a space in a proportional-pitch alphanumeric character string, characterized in that when a character rectangle interval ratio, which is a ratio of character rectangle intervals, is larger than a predetermined threshold value, it is determined that there is a space between corresponding characters.