JP2006276917A - Document processing device and document processing method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a technique for preventing missing of information generated by translating it into another language without giving a user trouble. <P>SOLUTION: A CPU 16 of a document processing device 1 tries to detect a phrase registered from one sentence of text data to become a candidate of translation. When the registered phrase is detected, the CPU 16 performs modification translation process to the one sentence concerned. In specific terms, in advance to the translation of this one sentence, the CPU 16 extracts a phrase which belongs to the predetermined range from this detected phrase from the one sentence concerned. Then the CPU 16 performs the processing to insert a character string enclosing the extracted phrase with parentheses immediately after the translation word of the phrase concerned concurrently with the processing for translating the one sentence concerned into Japanese. The above translation processing is repeatedly carried out also to the subsequent sentences, and same processing is performed to the whole sentence contained in the text data. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a technique for executing translation processing on document data, and more particularly to a technique for preventing loss of information caused by translation processing.

In recent years, with the internationalization of economic activities, a method for translating documents in a short time and at a low cost is desired, and a machine translation system is attracting attention from this viewpoint.
There are various methods for inputting document data to be translated into this machine translation system. One of them is to optically read a document to be translated, and to read characters from the read image data. There has been proposed a method of inputting document data by recognizing (see, for example, Patent Document 1).

In the translation copying apparatus disclosed in Patent Document 1, first, an original image is read by an image reading unit, and is discriminated into pattern information and character information by an identification unit. Of the identified information, the character information is recognized for each character by the translation means, and the content is translated into another language and output together with the original image. Therefore, according to this translation and copying apparatus, a hard copy obtained by translating the character information into another language can be obtained simply by copying the document to be translated.
Japanese Patent Laid-Open No. 62-154845

However, simply translating a document to be translated as in the above translation / copying apparatus may fail to translate the information contained in the word before translation without omission. For example, the phrase “Mr. Tanaka” or “Ms. Matsui” is translated into Japanese. If these words are simply translated into Japanese, they will be translated into Japanese, “Tanaka-san” and “Matsui-san” respectively. From these Japanese, “Tanaka-san”, “Matsui-san” Cannot know if she is male or female. This is because information for judging the gender is lost when translated from English to Japanese.
Conversely, when translating the words “Mr. Tanaka” and “Mr. Matsui” into English, they are translated into English such as “Mr./Ms. Tanaka” and “Mr./Ms. Matsui”. This is because the gender cannot be determined from the phrases “Mr. Tanaka” and “Mr. Matsui” to be translated.

The present invention has been made in view of the above-described circumstances, and an object of the present invention is to provide a technique for preventing loss of information caused by translation into another language without bothering the user. It is another object of the present invention to provide a technique for realizing accurate translation when translating from a phrase in another language into a phrase that lacks information by being translated into another language.
In the present specification, words such as “Mr.” and “Ms.” described above are referred to as semantically degenerate words.

  In order to solve the above-mentioned problem, the present invention includes a storage unit that stores a search character string for search in advance, an input unit that inputs a character string to be translated, and a character string input from the input unit. , A search character string detection means for detecting a character string that matches a search character string stored in advance in the storage means, and a character string input by the input means, and a search by the search character string detection means When a character string is detected, when translating the character string, a translation unit that inserts the search character string in the vicinity of the translated word of the character string, and a character string translated by the translation unit Provided is a document processing apparatus having output means for outputting.

  According to the present invention, a storage means for storing a search character string for search, an input means for inputting a character string to be translated, and a character string input from the input means are stored in advance in the storage means. A search character string detecting unit that detects a character string that matches the search character string, and when the search character string is detected by the search character string detection unit, the search character string is included in a predetermined range from the character string A replacement character string recognizing means for recognizing a character string to be read as a replacement character string for a phrase at a predetermined position with respect to the search character string, and translating the character string input from the input means, If there is a replacement character string recognized by the character string recognition means, a translation means that uses the replacement character string as a translated word of the phrase, and an output means for outputting the character string translated by the translation means; To provide a document processing apparatus having.

  According to the present invention, the document processing device stores the search process from the storage process for storing the search character string, the input process for inputting the character string to be translated, and the character string input in the input process. A search character string detection process for detecting a character string that matches the search character string stored in the process, a character string input in the input process is translated, and a search character string is detected in the search character string detection process In this case, when translating the character string, a translation process for inserting the search character string in the vicinity of the translated word of the character string and an output process for outputting the character string translated in the translation process are performed. Provide a document processing method to be executed.

  According to the present invention, the document processing device stores the search process from the storage process for storing the search character string, the input process for inputting the character string to be translated, and the character string input in the input process. A search character string detecting process for detecting a character string that matches the search character string stored in the search character string, and when a search character string is detected in the search character string detection process, A replacement character string recognition process for recognizing a character string included in the range as a replacement character string for a phrase at a predetermined position with respect to the search character string, and translating the character string input in the input process, If there is a replacement character string recognized in the replacement character string recognition process, a translation process using the replacement character string as a translated word of the phrase, and a translation in the translation process To provide a document processing method for performing an output process for outputting the string.

Hereinafter, an embodiment of the present invention will be described with reference to the drawings.
(1) Configuration FIG. 1 is a block diagram showing a configuration of a document processing apparatus 1 according to the first embodiment of the present invention.
The document processing apparatus 1 is operated by a user to input an instruction of the user, an operation unit 11, and a document set on a document table (not shown) of the document processing apparatus 1 is optically read and input as image data. Scanner 12, RAM 13 for temporarily storing image data, image of image data stored in RAM 13 is formed on paper, printing section 14 for discharging the paper as a document to outside document processing apparatus 1, and data stored And a CPU 16 for controlling each part of the document processing apparatus 1.

  The nonvolatile memory 15 includes a control program P11 that is used when the CPU 16 controls each unit of the document processing apparatus 1, a copy translation program P12 that is used to cause the document processing apparatus 1 to execute copy translation processing described later, and a semantic degenerate word list LST. And a dictionary D for translation.

  FIG. 2 is a diagram illustrating an example of the semantic degenerate word list LST. As shown in FIG. 2, meaning-reduced words are registered in advance in the meaning-reduced word list LST by the user, and are referred to by the CPU 16 when a copy translation process described later is executed.

When power is supplied from a power source (not shown), the CPU 16 reads the control program P11 from the nonvolatile memory 15 and executes it. As a result, the CPU 16 can control each part of the document processing apparatus 1.
In this state, the CPU 16 reads the copy translation program P12 from the non-volatile memory 15 and executes it when a translation instruction is input from the user using the operation unit 11. As a result, the document processing apparatus 1 executes the following copy / translation process.
First, the CPU 16 optically reads a document set on the document table of the document processing apparatus 1 by the scanner 12 and inputs it as image data, and writes the image data in the RAM 13. Next, the CPU 16 generates text data from the image data stored in the RAM 13 by OCR (Optical Character Recognition). Next, the CPU 16 processes the text data by translating the generated text data one sentence at a time. At this time, if the text data includes a word or phrase registered in the semantic degenerate word list LST, a processed translation process for accurately reflecting the information of the detected word or phrase in the translated sentence is performed. Executed. Next, the CPU 16 generates image data from the processed text data, forms an image of the image data on a sheet using the printing unit 14, and discharges the sheet to the outside of the document processing apparatus 1.

(2) Operation (2-1) Operation example 1
Next, a case where a document written in English is translated into Japanese using the document processing apparatus 1 according to the present embodiment will be described.
In the following description, it is assumed that the contents shown in FIG. 2 are registered in the semantic degenerate word list LST.

  First, when the user sets an English document shown in FIG. 3 on the manuscript table of the document processing apparatus 1 and inputs an instruction for translation using the operation unit 11, the CPU 16 starts a copy translation process for the document.

FIG. 4 is a flowchart showing a copy translation process according to this operation example.
First, the CPU 16 optically reads a document set on the platen by the scanner 12, inputs it as image data, and writes this image data in the RAM 13 (step SA1). Next, the CPU 16 generates text data by OCR from the image data stored in the RAM 13 (step SA2).

Next, the CPU 16 performs a translation process on the generated text data (step SA3). In this translation process, the dictionary D stored in the non-volatile memory 15 is referred to, and the text data is translated sentence by sentence. Hereinafter, this translation processing will be described with reference to FIG.
First, the CPU 16 tries to detect a word / phrase registered in the semantic degenerate word list LST from one sentence to be translated (step SB1). Since the document shown in FIG. 4 includes “Mr.” and “Ms.” registered in the semantic degenerate word list LST, the CPU 16 detects “Mr.” and “Ms.”. (Step SB1: YES). When a semantically degenerate word is detected, the CPU 16 executes a processed translation process on the sentence (step SB2). Specifically, first, the CPU 16 extracts “Mr.” and “Ms.” and words / phrases belonging to a predetermined range from these words / phrases before translation of the sentence / sentences. This predetermined range can be set by the user, and in this embodiment, it is assumed that the predetermined range is set to a range including the detected word and one word following the word. Therefore, in the case of the document shown in FIG. 4, “Mr. Tanaka” and “Ms. Matsui” are extracted.
Next, in parallel with the process of translating the sentence into Japanese, the CPU 16 converts the extracted strings “Mr. Tanaka” and “Ms. Matsui” into parentheses for each translated word of the phrase. The process of inserting immediately after is executed. The insertion position of the word may be immediately before the translated word or below the translated word. In short, it may be in the vicinity of the translated word.

On the other hand, when a word registered in the semantically degenerate word list LST is not detected from one sentence to be translated (step SB1: NO), a normal translation process is executed on the one sentence (step SB3). .
The above translation processing is repeatedly executed for subsequent sentences. When the same processing is executed for all sentences included in the text data (step SB4: YES), the CPU 16 ends the translation processing.

  Next, the CPU 16 generates image data from the text data obtained as a result of the translation process, and overwrites the RAM 13 (step SA4, FIG. 4). As a result, the image data stored in the RAM 13 becomes the image data of the document shown in FIG. Next, the CPU 16 forms an image of the generated image data on a sheet using the printing unit 14, and discharges the sheet to the outside of the document processing apparatus 1 (step SA5).

  As described above, according to the document processing apparatus 1 of the present embodiment, when a semantically degenerate word is translated into another language, the semantically degenerate word is automatically inserted immediately after the translated word. The loss of information caused by the translation process can be prevented without trouble.

(2-2) Operation example 2
Next, a case where a document written in Japanese is translated into English using the document processing apparatus 1 according to the present embodiment will be described.
In the following description, it is assumed that the contents shown in FIG. 2 are registered in the semantic degenerate word list LST.

  First, when the user sets a Japanese document shown in FIG. 7 on the manuscript table of the document processing apparatus 1 and inputs a translation instruction using the operation unit 11, the CPU 16 starts a copy translation process for the document. . As shown in FIG. 7, it is assumed that the translated word of the word is appended in parentheses to the Japanese word or phrase described in this document. However, the symbols surrounding the translated word are not limited to the parentheses as shown in FIG. 7, and may be other symbols.

Hereinafter, the copy translation process according to this operation example will be described with reference to FIG.
First, the CPU 16 optically reads a document set on the platen by the scanner 12, inputs it as image data, and writes this image data in the RAM 13 (step SA1). Next, the CPU 16 generates text data by OCR from the image data stored in the RAM 13 (step SA2).

Next, the CPU 16 performs a translation process on the generated text data (step SA3). In this translation process, the dictionary D stored in the non-volatile memory 15 is referred to, and the text data is translated sentence by sentence. Hereinafter, this translation processing will be described with reference to FIG.
First, the CPU 16 tries to detect a word / phrase registered in the semantic degenerate word list LST and enclosed in parentheses from a sentence to be translated (step SB1). Since the document shown in FIG. 7 includes “Mr.” and “Ms.” registered in the semantic degenerate word list LST, the CPU 16 detects “Mr.” and “Ms.”. (Step SB1: YES). When a semantically degenerate word is detected in this way, the CPU 16 executes a processed translation process for the sentence (step SB2). Specifically, first, the CPU 16 performs a process of extracting a phrase included in parentheses surrounding the detected phrase for each detected phrase. In the case of the document shown in FIG. 7, “Mr. Tanaka” and “Ms. Matsui” are extracted. Next, the CPU 16 executes a process of replacing these extracted words / phrases with words / phrases located at predetermined positions from the respective words / phrases. In the case of the document shown in FIG. 7, “Mr. Tanaka” is first replaced with “Mr. Tanaka”, and “Ms. Matsui” is replaced with “Ms. Matsui”. When the replacement process is completed, the CPU 16 performs a process of translating the sentence into English.

On the other hand, when a word registered in the semantically degenerate word list LST is not detected from one sentence to be translated (step SB1: NO), a normal translation process is executed on the one sentence (step SB3). .
The above translation processing is repeatedly executed for subsequent sentences. When the same processing is executed for all sentences included in the text data (step SB4: YES), the CPU 16 ends the translation processing.

  Next, the CPU 16 generates image data from the text data obtained as a result of the translation process, and overwrites the RAM 13 (step SA4, FIG. 4). As a result, the image data stored in the RAM 13 becomes the image data of the document shown in FIG. Next, the CPU 16 forms an image of the generated image data on a sheet using the printing unit 14, and discharges the sheet to the outside of the document processing apparatus 1 (step SA5).

  As described above, according to the document processing apparatus 1 of the present embodiment, when a document written in another language is translated into a semantic degenerate word, the semantic degenerate word described in a predetermined position of the document, its translation sentence, Since the translation process is automatically performed after replacing the information, it is possible to prevent the loss of information caused by the translation process without bothering the user.

(3) Modifications Embodiments of the present invention may be modified as described below.
For example, in the description of the above-described embodiment, the description has been made assuming that the semantic degenerate word is registered in the semantic degenerate word list LST in advance. However, the document processing apparatus 1 is provided with a copy registration program and the semantic degenerate word is described. The semantically degenerate word may be registered by copying the document using the document processing apparatus 1. Hereinafter, this registration processing method will be described.
First, when a copy instruction is input using the operation unit 11, the CPU 16 of the document processing apparatus 1 reads out and executes a copy registration program from the nonvolatile memory 15. In this copy registration process, the CPU 16 optically reads a document set on the document table of the document processing apparatus 1 by the scanner 12, inputs it as image data, and writes it in the RAM 13. In the document read here, it is assumed that a semantic degenerate word that the user desires to register is written according to a predetermined format. Next, the CPU 16 uses the printing unit 14 to form an image of the image data on a sheet by extracting the semantically degenerate word from the image data stored in the RAM 13 and registering it in the semantic degenerate word list LST. Thus, processing for discharging the paper document processing apparatus 1 to the outside is executed in parallel.
As a result, by simply copying the document, the semantically degenerate word can be registered in the semantically degenerate word list LST.

  Further, in the above-described embodiment, the application example of the present invention has been described by taking the case of translating from English to Japanese and vice versa. However, the present invention is translated by the present invention. The translated word and the translated word are not limited to these languages, and may be any language.

The document processing device 1 is not limited to a device that functions as a copying machine, but a scanner device that optically reads a document and outputs image data, a facsimile device that optically reads a document and transmits image data, or It may be a printer that inputs image data of a document and prints the document.
Further, the document processing apparatus 1 may be an apparatus that functions as a translation apparatus that inputs text data of a document and sends back translated text data.

  Further, in the description related to the translation processing of the operation example 1 described above, it is assumed that the detected semantic degenerate word and the phrase belonging to the predetermined range are extracted from the word and inserted immediately after the translated word of the semantic degenerate word. Although the explanation is given, the extracted and inserted words may be only semantically degenerate words. For example, taking operation example 1 as an example, only "Mr." and "Ms." registered in the semantically degenerate word list LST are extracted and inserted immediately after "san", which is the translated word of these phrases. Will be. As a result, one sentence after translation becomes “Mr. Tanaka (Mr.) and Ms. Matsui (Ms.)”.

Further, in the description related to the translation processing of the above-described operation example 2, the detected semantically degenerate word and the phrase included in parentheses surrounding the phrase are extracted, and the extracted phrase and the phrase belonging to a predetermined range from the phrase In the above description, the document processing apparatus 1 is registered with the word / phrase group to be replaced as a replacement word list, and the replacement process is performed by referring to the replacement word / phrase list in the subsequent translation processing. Processing may be executed. FIG. 9 is a diagram illustrating an example of the replacement phrase list.
According to this configuration, for a word / phrase that has been replaced once, the replacement word / phrase is specified by referring to the replacement word / phrase list, so that the translated word of the word / phrase after the Japanese word / phrase as shown in FIG. Accurate translation can be realized without adding.

It is a block diagram which shows the structure of the document processing apparatus 1 which concerns on one Embodiment of this invention. It is a figure which shows an example of the semantic degenerate word list LST in the document processing apparatus. 3 is a diagram illustrating an example of a document input to the document processing apparatus 1. FIG. 4 is a flowchart showing a copy / translation process performed by the document processing apparatus 1; It is a flowchart which shows the translation process which the document processing apparatus 1 performs. 3 is a diagram illustrating an example of a document formed by the document processing apparatus 1. FIG. 3 is a diagram illustrating an example of a document input to the document processing apparatus 1. FIG. 2 is an example of a document formed by the document processing apparatus 1; It is a figure which shows an example of the phrase replacement list which concerns on one modification of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... Document processing apparatus, 11 ... Operation part, 12 ... Scanner, 13 ... RAM, 14 ... Printing part, 15 ... Nonvolatile memory, 16 ... CPU, D ... Dictionary, LST ... Semantic degenerate word list, P11 ... Control program, P12: Copy translation program.

Claims (4)

Storage means for storing a search character string for search in advance;
An input means for inputting a character string to be translated;
Search character string detection means for detecting a character string that matches a search character string stored in advance in the storage means from a character string input from the input means;
When the character string input by the input means is translated and the search character string is detected by the search character string detection means, when the character string is translated, the position near the translated word of the character string A translation means for inserting the search character string into
A document processing apparatus comprising: output means for outputting a character string translated by the translation means. Storage means for storing a search character string for search;
An input means for inputting a character string to be translated;
Search character string detection means for detecting a character string that matches a search character string stored in advance in the storage means from a character string input from the input means;
When a search character string is detected by the search character string detection means, the search character string and a character string included in a predetermined range from the character string are replaced with a phrase at a predetermined position with respect to the search character string A replacement character string recognition means for recognizing as a character string,
Translating the character string input from the input means, and if there is a replacement character string recognized by the replacement character string recognition means, a translation means using a replacement character string as a translated word of the phrase;
A document processing apparatus comprising: output means for outputting a character string translated by the translation means. The document processing device
A storage process for storing a search character string for search;
An input process for inputting a character string to be translated;
A search character string detection step of detecting a character string matching the search character string stored in the storage step from the character string input in the input step;
When the character string input in the input process is translated and the search character string is detected in the search character string detection process, when the character string is translated, the position near the translated word of the character string The translation process of inserting the search string into
And a document processing method for executing an output process of outputting a character string translated in the translation process. The document processing device
A storage process for storing a search character string for search;
An input process for inputting a character string to be translated;
A search character string detection step of detecting a character string matching the search character string stored in the storage step from the character string input in the input step;
When a search character string is detected in the search character string detection process, the search character string and a character string included in a predetermined range from the character string are replaced with a phrase at a predetermined position with respect to the search character string A replacement character string recognition process to be recognized as a character string;
Translating the character string input in the input process and, if there is a replacement character string recognized in the replacement character string recognition process, a translation process using a replacement character string as a translated word of the phrase;
And a document processing method for executing an output process of outputting a character string translated in the translation process.

Images