JP2007034430A - Translation device, translation method, and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a technology which can make the layout of a translated sentence identical to that of an original sentence when a document generated with a certain language is translated with a different language. <P>SOLUTION: A translation device 1 determines, for each unit area which constitutes an original sentence, whether a translated sentence corresponding to an original sentence arranged in a unit area can be arranged with the same layout as the original sentence in the unit area. When the result of the determination is that the translated sentence cannot be arranged with the same layout as the original sentence in the unit area, the translation device extracts words and phrases, which satisfy preliminarily determined conditions, from the translated sentence as the abbreviation-object words and phrases, and then determines abbreviation words obtained by abbreviating part of a character string constituting the abbreviation-object words and phrases. Then, the translation device generates a translated sentence in which the abbreviation-object words and phrases are converted to abbreviation words, generates the image of the translated sentence with the same layout as the original sentence, and generates images representing the definitions of the abbreviation words. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a technique for translating a document into a document of a different language.

  If a sentence written in one language is translated into a sentence in another language, the number of characters in the sentence may increase. For example, translating Japanese kanji-kana mixed sentences into English generally increases the number of characters. FIG. 4 is an example of the original text written in Japanese. FIG. 5 is an example of a translation sentence obtained by translating the original sentence into English. As shown in this example, if the number of characters in the translated sentence is larger than the number of characters in the original sentence, the translated sentence image may not be generated with the same layout as the original sentence. Here, the layout means the arrangement of sentences and diagrams on the page, character size, character spacing, line spacing, and so on.

For example, if an image of a translated sentence is generated with the same character size, character spacing, and line spacing as the original sentence, the length of the sentence on the external appearance becomes longer. When the length of the appearance of a sentence becomes long, the impression of the appearance of the document changes because the positions of chapters, sections, paragraph breaks, charts and the like may change. In addition, the number of pages may increase. However, it is desirable that the impressions of the appearance of the original text and the translated text are similar. It is also desirable to avoid an increase in the number of pages.
On the other hand, if the character size of the translated sentence is reduced, the length of the appearance of the sentence can be shortened. However, if the character size is reduced, the text becomes difficult to read. Even if means such as reducing the character spacing or the line spacing are used, the text becomes difficult to read.
As described above, when translating a document into another language, it is impossible with the conventional technique to always make the translated sentence layout the same as the original sentence.

By the way, a technique for creating an abbreviation by deleting a part of a character string constituting a phrase and creating a sentence using the abbreviation is known. For example, in the technique disclosed in Patent Document 1, a word string corresponding to a predetermined condition is extracted, and a part of characters included in the extracted word string is deleted according to a certain rule. Create a word. For example, if a predetermined number or more of a specific part of speech such as a noun is continuous and there are a predetermined number or more of the same word string in the sentence, an abbreviation is created using the first character of the word included in the word string . Then, a sentence in which this word string is replaced with an abbreviation is generated. As a result, frequently occurring long phrases are automatically converted into abbreviations, so that an easy-to-read sentence can be created without increasing the burden on the user.
JP 11-328166 A

If the technique disclosed in Patent Document 1 is used, the number of characters in the sentence can be reduced. However, this technique only reduces the number of characters in a word string that meets a predetermined condition, and thus cannot control the length of the appearance of a sentence. Therefore, with this technique, the layout of the translated sentence cannot always be the same as the original sentence.
The present invention has been made under the above-described background, and provides a technique capable of making the layout of the translated sentence the same as that of the original sentence when translating a document created in a language into a document in a different language. With the goal.

  In order to solve the above-described problems, the present invention relates to a dictionary that stores words having the same meaning in different languages in association with each other, an input unit that inputs an image of a document in which the original text is described, and an input unit that uses the input unit. An image analyzing unit that analyzes a layout of a document represented by the displayed image and recognizes a unit area constituting the document, and a character that recognizes a character string of the original text included in the image input by the input unit Recognizing means; translation means for extracting a translation word corresponding to the character string recognized by the character recognizing means from the dictionary; generating a translated sentence; and for each unit area recognized by the image analyzing means, Determining means for determining whether or not a translated sentence corresponding to the original sentence arranged in the area can be arranged in the unit area in the same layout as the original sentence; and as a result of the determination by the determining means, the translation The sentence When it is impossible to arrange in the unit area in the same layout, an extraction unit that extracts a phrase that satisfies a predetermined first condition from the translated sentence as an abbreviation target phrase, and extracted by the extraction unit An abbreviation determination means for determining an abbreviation in which a part of a character string constituting the abbreviation target phrase is omitted, and an abbreviation target phrase extracted by the extraction means is converted into an abbreviation determined by the abbreviation determination means A conversion means for generating a translation sentence, and an image representing the definition of the abbreviation determined by the abbreviation determination means while generating an image of the translation sentence generated by the conversion means in the same layout as the original sentence There is provided a translation apparatus comprising: an image generating means for generating; and an output means for outputting an image generated by the image generating means.

  In order to solve the above-described problem, the present invention analyzes an input step of inputting an image of a document in which the original text is described, and a layout of the document represented by the image input in the input step, An image analysis step for recognizing a unit area constituting the document, a character recognition step for recognizing a character string of the original text included in the image input in the input step, and a character string recognized in the character recognition step For each unit region recognized in the image analysis step, a translation step that extracts a corresponding translated word from a dictionary that stores words having the same meaning in different languages in association with each other and generates a translated sentence. A determination step for determining whether or not a translated sentence corresponding to the arranged original sentence can be arranged in the unit area in the same layout as the original sentence; and As a result of the determination, when the translated sentence cannot be arranged in the unit area in the same layout as the original sentence, a phrase satisfying a first predetermined condition is extracted from the translated sentence as an abbreviation target phrase. An abbreviation determination step for determining an abbreviation obtained by omitting a part of the character string constituting the abbreviation target word / phrase extracted in the extraction step; an abbreviation target word / phrase extracted in the extraction step; A conversion step for generating the translated sentence converted into the abbreviation determined in the determination step, and an image of the translation sentence generated in the conversion step is generated in the same layout as the original sentence, and in the abbreviation determination step An image generation step for generating an image representing the definition of the abbreviation determined; and an output step for outputting the image generated in the image generation step. To provide a translation method comprising Rukoto.

  In order to solve the above-described problem, the present invention provides a computer apparatus in which a dictionary storing words having the same meaning in different languages in association with each other, and an input unit for inputting an image of a document in which the original text is described. Analyzing the layout of the document represented by the image input by the input means, recognizing the unit area constituting the document, and the original text included in the image input by the input means Character recognition means for recognizing a character string, translation means for extracting a translation word corresponding to the character string recognized by the character recognition means from the dictionary and generating a translation sentence, and a unit recognized by the image analysis means A determination unit that determines whether a translated sentence corresponding to an original sentence arranged in the unit area can be arranged in the unit area for each area in the same layout as the original sentence; and Judgment As a result, when the translated sentence cannot be arranged in the unit area in the same layout as the original sentence, an extraction means for extracting from the translated sentence a word that satisfies a predetermined first condition as an abbreviation target phrase An abbreviation determination means for determining an abbreviation obtained by omitting a part of a character string constituting the abbreviation target phrase extracted by the extraction means, and an abbreviation determination means for the abbreviation target phrase extracted by the extraction means A conversion unit that generates a translated sentence converted into the abbreviation determined in step (b), an image of the translation sentence generated by the conversion unit is generated in the same layout as the original sentence, and the abbreviation determination unit determines An image generation means for generating an image representing the definition of the abbreviation and a program for functioning as an output means for outputting the image generated by the image generation means.

  According to the above translation apparatus, translation method, or program, for each unit area constituting the original sentence, the translated sentence corresponding to the original sentence arranged in the unit area is arranged in the unit area in the same layout as the original sentence. It is determined whether or not As a result of the determination, if the translated sentence cannot be arranged in the unit area in the same layout as the original sentence, a phrase satisfying a first predetermined condition is extracted from the translated sentence as an abbreviation target phrase. Next, an abbreviation in which a part of the character string constituting the abbreviation target phrase is omitted is determined. Next, a translation sentence in which the abbreviation target phrase is converted into an abbreviation is generated, an image of the translation sentence is generated with the same layout as the original sentence, and an image representing the definition of the abbreviation is generated.

  According to the present invention, when a document created in a language is translated into a document in a different language, the layout of the translated sentence can be made the same as the original sentence.

Embodiments of the present invention will be described below with reference to the drawings.
<First Embodiment>
<Configuration>
The translation apparatus 1 has a translation function for translating an input original sentence to obtain a target language translation sentence, and a copying function for reading a manuscript and creating a duplicate.
First, the hardware configuration of the translation apparatus 1 will be described. FIG. 1 is a diagram illustrating a hardware configuration of a translation apparatus 1 according to an embodiment of the present invention.
The translation apparatus 1 includes a control unit 4 including a CPU (Central Processing Unit) 44, a ROM (Read Only Memory) 45, a RAM (Random Access Memory) 46, and the like (all not shown), and is stored in the ROM 45. Each part of the translation apparatus 1 is controlled by the CPU 44 executing an OS (Operating System) program.

The storage unit 5 is a nonvolatile memory such as a hard disk device. The storage unit 5 stores a program describing procedures such as reading, translating, and outputting a document.
The instruction input unit 41 includes a keyboard 40 including a numeric keypad and a start button, and a display unit 39 including a liquid crystal panel having a touch panel function. The user can input an instruction to the translation apparatus 1. In addition, when an abnormality or the like occurs in the translation apparatus 1, information regarding the state of the translation apparatus 1 can be displayed on the display unit 39.
A sheet 10 for forming an image is accommodated in the sheet feed tray 9. When the user inputs an image formation instruction using the instruction input unit 41, the paper feed roller 33 is driven to rotate, and the sheets 10 are sent out from the paper feed tray 9 one by one. The sheet 10 sent out from the paper feed tray 9 is conveyed along a conveyance path 36 by roller pairs 34, 35, and 37.

  The image input unit 12 is a scanner device that optically reads a document and generates image data. The original placed on the platen glass 2 is irradiated with light from the light source 13, and the reflected light is processed by the optical system 3. The reflected light is received by the light receiving unit 17 via the mirrors 14, 15 and 16. Then, the image processing unit 18 converts the reflected light into an electrical signal, and generates image data composed of each color of yellow, magenta, cyan, and black.

The image forming unit 6 includes image forming engines 7Y, 7M, 7C, and 7K, a transfer belt 8, and the like.
The image forming engines 7Y, 7M, 7C, and 7K form toner images of yellow (Y), magenta (M), cyan (C), and black (K), respectively. Since the configuration of each image forming engine is common, only the image forming engine 7Y will be described here.
The image forming engine 7Y includes a charging device 21Y, an exposure device 19Y, a developing device 22Y, a cleaner 24Y, and the like around a photosensitive drum 20Y as an image carrier on which an electrostatic latent image is formed.

The charging device 21Y charges the surface of the photosensitive drum 20Y that is rotationally driven in the direction of arrow A to a predetermined potential.
The exposure device 19Y is a ROS (Raster Output Scanner) that irradiates a photosensitive drum 20Y charged to a predetermined potential with an exposure beam LB based on image data. The exposure device 19Y emits a laser beam based on image data from a semiconductor laser (not shown), and forms an electrostatic latent image on the surface of the photoreceptor drum 20Y by deflecting and scanning the laser beam. On the surface of the photoreceptor drum 20Y, the potential of the portion irradiated with the laser light is reduced to a predetermined level due to the photoconductivity of the photoreceptor drum 20Y. As described above, when the surface potential of the photosensitive drum 20Y changes, an electrostatic latent image based on the image data is formed on the surface of the photosensitive drum 20Y.

The developing device 22Y is a device that visualizes the electrostatic latent image formed on the surface of the photosensitive drum 20Y. Toner (charged color material) is supplied from the toner tank 23Y, and a toner image is obtained by reversal development of the electrostatic latent image with toner charged to the same polarity as that of the photosensitive drum 20Y.
The transfer belt 8 is stretched around rollers 26, 27, 28, and 29, and is circulated and driven in the direction of arrow B. The photosensitive drum 20Y is in pressure contact with the transfer belt 8 positioned below the photosensitive drum 20Y, and the toner image formed as described above is transferred to the transfer belt 8.
The cleaner 24Y is a device that removes the toner remaining on the photosensitive drum 20Y.

  The above is the configuration of the image forming engine 7Y. In the image forming engines 7M, 7C, and 7K, toner images corresponding to the respective colors are formed and transferred onto the transfer belt 8 in an overlapping manner. Hereinafter, when it is not necessary to distinguish the image forming engines 7Y, 7M, 7C, and 7K, they are simply referred to as the image forming engine 7. Similarly, for other components, the Y, M, C, and K notations are omitted when it is not necessary to distinguish between Y, M, C, and K.

The sheet 10 sent out from the paper feed tray 9 onto the conveyance path 36 enters a nip portion formed by the transfer belt 8 and the transfer roller 30 and is pressed against the transfer belt 8. The toner image is transferred to the surface of the sheet 10 by this pressure contact force and electrostatic attraction force.
The sheet 10 on which the toner image is transferred is guided to the fixing device 11 by the roller pair 31. In the fixing device 11, the sheet 10 is pressed and heated to fix the toner image on the sheet 10. The sheet 10 on which the image is formed in this manner is discharged to the paper discharge tray 32.

Next, the functional configuration of the translation apparatus 1 will be described. FIG. 2 is a diagram illustrating a functional configuration of the translation apparatus 1. These functions are realized by the CPU 44 executing the program stored in the storage unit 5.
The dictionary 201 stores words having the same meaning in different languages in association with each other, and is stored in the storage unit 5. The dictionary 201 stores various languages such as Japanese, English, and Chinese. Further, the dictionary 201 stores information indicating the type of part of speech of the word in association with each word.

The abbreviation dictionary 202 stores abbreviations in which a part of a character string constituting a phrase is omitted in association with the phrase, and is stored in the storage unit 5. The abbreviation dictionary 202 also stores data corresponding to various languages such as Japanese, English, and Chinese.
FIG. 6 is a diagram illustrating an example of the abbreviation dictionary 202. In the figure, an example of an English abbreviation dictionary is shown. For example, “NG” is stored in association with the abbreviation of the original word “next-generation”. The abbreviation dictionary 202 may store abbreviations arbitrarily determined in advance, or may store abbreviations generated by the abbreviation generator 107 described later.

The input unit 101 inputs an image of a document in which the original text is described. Specifically, the input unit 101 inputs an image of a document using the image input unit 12 and generates document image data representing the image of the document.
The image analysis unit 102 analyzes the layout of the document represented by the image input by the input unit 101 and recognizes unit areas constituting the document. Here, the layout means the arrangement of sentences and diagrams on the page, character size, character spacing, line spacing, and so on. Through the layout analysis, a text area, a chart area, a picture area, and the like constituting the original are distinguished and recognized. Furthermore, chapters, sections, paragraphs, etc. in the text area, and cell divisions in the table are recognized. Each area delimited by these delimiters is recognized as a unit area. For example, in the example of FIG. 4, the line from “1. Introduction” to the line immediately before “2.XML features” is recognized as one unit quantity region.

  The character recognition unit 103 recognizes a character string of the original text included in the image input by the input unit 101. The storage unit 5 stores a known OCR (Optical Character Recognition) program, and character recognition is performed by the CPU 44 executing the OCR program.

  The translation unit 104 translates the original sentence included in the image input by the input unit 101 into a sentence described in the translation destination language to generate a translation sentence. A known method is used to generate a translation. For example, the original text is divided into morphemes using a morphological analysis technique, the syntax of the original text is analyzed using a syntax analysis technique and a semantic analysis technique, and a syntax tree is created, which is converted into a syntax tree in the target language. Then, the translation corresponding to each morpheme is extracted from the dictionary 201 and applied to the syntax tree to generate a translated sentence.

  For each unit region recognized by the image analysis unit 102, the determination unit 105 can arrange the translated sentence corresponding to the original sentence arranged in the unit area in the unit area in the same layout as the original sentence. Determine whether or not. For example, in the unit area following “1. Introduction” in the example of FIG. 4, the number of translated text lines (18 lines) exceeds the number of original text lines (14 lines) as shown in the example of FIG. Yes. Therefore, in this example, it is determined that the translated sentence cannot be arranged in the unit area in the same layout as the original sentence.

If the result of determination by the determining unit 105 is that the translated sentence cannot be arranged in the unit area in the same layout as the original sentence, the extracting unit 106 selects a phrase that satisfies the first condition that is determined in advance. As extracted from the translation. Here, the predetermined first condition is, for example, a case where the number of appearances of a noun or a noun phrase having more characters than the predetermined upper limit exceeds the predetermined upper limit. . FIG. 6 shows an example of the abbreviation dictionary 202. For example, the original word “next-generation” has 15 characters. When the upper limit of the number of characters is set to 10, for example, “next-generation” is extracted as an abbreviation target phrase. The upper limit value of the number of appearances is an arbitrary value and may be 1.
In addition to the above, any condition can be used as the condition for extracting the omitted words / phrases in the extracting means 106. For example, idioms may be extracted as abbreviation target phrases with priority over words. Alternatively, the abbreviation target phrases may be extracted so that the number of abbreviations used in the translated sentence is minimized.

The abbreviation generation means 107 generates abbreviations in which a part of a character string constituting a phrase is omitted according to a predetermined rule. An arbitrary method can be used as the method for generating the abbreviation. Examples of abbreviation generation methods are listed below.
<Method 1> When the abbreviation target phrase is a word, an abbreviation is generated using a predetermined number of characters from the beginning of the abbreviation target phrase. For example, when the abbreviation target phrase is “specification”, the abbreviation “spec” is generated by using the first four characters.
<Method 2> When the abbreviation target phrase is a word, the abbreviation is generated by arranging the first character of the abbreviation target phrase and the consonant following the first character in the order of appearance. For example, when the abbreviation target phrase is “source”, an abbreviation “src” is generated.
<Method 3> In Method 2, only the consonant that appears first among the consonants following the first character is used. For example, when the abbreviation target phrase is “green”, an abbreviation “gr” is generated.

<Method 4> When an abbreviation target phrase is a idiom (a phrase composed of two or more words), an abbreviation is generated by arranging the first characters of the words constituting the abbreviation target phrase in the order of appearance. For example, when the abbreviation target phrase is “markup language”, an abbreviation “ML” is generated.
<Method 5> When the abbreviation target phrase is a idiom, the first character of each word excluding the word that appears last among the words that constitute the abbreviation target phrase, and all the characters that constitute the word that appears last Abbreviations are generated by arranging them in order of appearance. For example, when the abbreviation target phrase is “target value”, an abbreviation “tvalue” is generated.

The abbreviation determination means 108 determines an abbreviation obtained by omitting a part of the character string constituting the abbreviation target phrase extracted by the extraction means 106. The abbreviation is determined using the abbreviation dictionary 202 and the abbreviation generator 107. First, abbreviations corresponding to the abbreviation target phrases extracted by the extraction unit 106 are extracted from the abbreviation dictionary 202. If the abbreviation corresponding to the abbreviation target phrase is not stored in the abbreviation dictionary 202, the abbreviation generation means 107 generates an abbreviation for the abbreviation target phrase.
In determining abbreviations, a translated sentence that fits in the same layout as the original sentence is not always generated in the first trial. For this reason, after the abbreviation determination means 108 once determines an abbreviation, it is determined using the determination means 105 whether or not a translation using this abbreviation fits in the same layout as the original sentence. As a result of the determination, if they do not fit in the same layout, an abbreviation is further generated by lowering the upper limit of the number of characters or the upper limit of the number of appearances (second condition), and a translation using this abbreviation is It is determined using the determination means 105 whether or not it fits in the same layout as the original text. The abbreviation determination means 108 repeats this series of processing until the translated sentence using the abbreviation comes into the same layout as the original sentence.

The conversion unit 109 generates a translated sentence obtained by converting the abbreviation target phrase extracted by the extraction unit 106 into the abbreviation determined by the abbreviation determination unit 108.
The image generation unit 110 generates an image of the translated sentence generated by the conversion unit 109 with the same layout as the original sentence, and generates an image representing the definition of the abbreviation determined by the abbreviation determination unit 108. An image representing the definition of the abbreviation is placed, for example, in the margin of the translated sentence. The image generation unit 110 generates document data including a text representing a translated sentence and an abbreviation definition and a document layout using a page description language.
The output unit 111 outputs an image based on the data generated by the image generation unit 110. Specifically, the document data generated by the image generation unit 110 is converted into raster data and supplied to the image forming unit 6. Then, an image representing the translated text is formed on the sheet surface and discharged to the paper discharge tray 32.

<Operation>
Next, the operation of the translation apparatus 1 will be described. Here, it is assumed that the translation apparatus 1 is powered on and the CPU 44 is executing a program. FIG. 3 is a diagram illustrating an operation flow of the translation apparatus 1. In addition, since the processing shown below is performed by the CPU 44 executing a program stored in the storage unit 5, in the following description, the main subject of the operation is the CPU 44.
First, in step A01, the CPU 44 uses the input unit 101 to read a paper document (original) in which the original text is described. FIG. 4 shows an example of a manuscript in which the original text (Japanese) is written. The user places the document so that the surface on which the original text is written is in contact with the platen glass 2, and presses the start button of the instruction input unit 41. Then, the image input unit 12 starts reading an image. In step A <b> 02, the CPU 44 generates image data representing the read original image and stores it in the storage unit 5.

Next, in step A03, the CPU 44 uses the image analysis unit 102 to analyze the layout of the document represented by the input image, and recognizes unit areas constituting the document. In the example of FIG. 4, a line from “1. Introduction” to a line immediately before “2. XML features” is recognized as one unit quantity region.
In step A04, the CPU 44 uses the character recognition means 103 to recognize the original character string contained in the input image.

In step A05, the CPU 44 uses the translation unit 104 to translate the original sentence included in the input image into a sentence described in the translation destination language to generate a translated sentence. FIG. 5 shows an example in which the original text (Japanese) shown in FIG. 4 is translated into English. In FIG. 4, the number of lines of the original text arranged in the unit area following “1. Introduction” is 14 lines, whereas in FIG. 5, the number of translated text lines of the original text is increased to 18 lines. is doing.
In step A06, using the determination unit 105, the CPU 44 can arrange, for each unit area, a translated sentence corresponding to the original sentence arranged in the unit area in the unit area in the same layout as the original sentence. It is determined whether or not. For example, in the unit area following “1. Introduction” in FIG. 4, the number of lines in the translated sentence exceeds the number of lines in the original sentence as shown in FIG. Therefore, in this example, it is determined that the translated sentence cannot be arranged in the unit area in the same layout as the original sentence (step A06: NO), and the process proceeds to step A07.

  In step A07, if the result of determination by the determination means 105 is that the translated sentence cannot be placed in the unit area in the same layout as the original sentence, the CPU 44 uses the extraction means 106 to determine a first predetermined value. A phrase satisfying the above condition is extracted from the translated sentence as an abbreviation target phrase. In the example of FIG. 5, if the upper limit of the number of characters is 13 and the upper limit of the number of appearances is 1, a total of four words “next-generation”, “markup language”, “specification”, and “electronic commerce” can be omitted. Extracted as a phrase.

  In step A08, the CPU 44 uses the abbreviation determination means 108 to determine an abbreviation in which a part of the character string constituting the abbreviation target phrase extracted by the extraction means 106 is omitted. In this example, as shown in FIG. 6, since the abbreviations of the above four terms are stored in the abbreviation dictionary 202, the abbreviations corresponding to each abbreviation target phrase are extracted from the abbreviation dictionary 202. . In this example, in step A09, the abbreviations of the above four words are stored in the abbreviation dictionary 202, so the abbreviations are not stored in the abbreviation dictionary 202 (step A09).

In step A10, the CPU 44 uses the conversion means 109 to convert the abbreviation target phrase into the abbreviation determined by the abbreviation determination means 108. Then, returning to step A06, the translation means corresponding to the original text arranged in the unit area can be arranged in the unit area in the same layout as the original text for each unit area using the determination unit 105. It is determined whether or not. FIG. 7 is a diagram showing an example of a translated sentence using abbreviations. As shown in the figure, the number of translated sentences is 14 by using abbreviations. Therefore, in this example, it is determined that the translated sentence corresponding to the original sentence arranged in the unit area can be arranged in the unit area with the same layout as the original sentence (step A06: YES). As a result of this determination, the CPU 44 proceeds to step A11 and uses the image generation means 110 to generate an image representing the translated sentence obtained by converting the abbreviation target phrase into an abbreviation with the same layout as the original sentence. Furthermore, an image representing the definition of the abbreviation is placed in the margin of the translated sentence image. Then, document data representing the generated image is generated and supplied to the output unit 111. The CPU 44 converts the document data into raster data using the output unit 44 and supplies the raster data to the image forming unit 6. Then, an image representing the translated text is formed on the sheet surface and discharged to the paper discharge tray 32.
The above is the operation flow of the translation apparatus 1.

  Next, an example of converting an abbreviation target phrase described in the table into an abbreviation will be described. FIG. 8 shows an example of a table written in Japanese. When this table is translated into a table written in English, the words placed in the “color” column will not fit within the same layout of the original text, that is, the corresponding cell frame, because the number of characters increases when translated into English. . Therefore, in this example, the words / phrases arranged in the “color” column are set as abbreviation target words / phrases and these are converted into abbreviations. FIG. 9 is a diagram illustrating an example of a table created using abbreviations. In this example, an abbreviation is created using two characters from the beginning of the abbreviation target phrase. “Color” is converted to “co”, “black” to “bl”, “yellow” to “ye”, and “green” to “gr”. Then, a character string representing the definition of each abbreviation is output in the margin (in this example, the upper side of the table). In this way, the translated sentence can be created with the same layout as the original sentence.

  As described above, the translation apparatus according to the present invention arranges, for each unit area constituting the original sentence, the translated sentence corresponding to the original sentence arranged in the unit area in the unit area in the same layout as the original sentence. It is determined whether or not it can be performed. As a result of the determination, if the translated sentence cannot be arranged in the unit area in the same layout as the original sentence, a phrase satisfying a first predetermined condition is extracted from the translated sentence as an abbreviation target phrase. Next, an abbreviation in which a part of the character string constituting the abbreviation target phrase is omitted is determined. Next, a translation sentence in which the abbreviation target phrase is converted into an abbreviation is generated, an image of the translation sentence is generated with the same layout as the original sentence, and an image representing the definition of the abbreviation is generated. Therefore, according to the present invention, when a document created in one language is translated into a document in a different language, the layout of the translated sentence can be made the same as the original sentence.

<Modification>
The present invention is not limited to the form described above, and can be implemented in various forms. For example, the embodiment described above can be modified as follows.

  In the embodiment described above, the abbreviation corresponding to the abbreviation target phrase is extracted from the abbreviation dictionary 202 when the abbreviation is determined, and the abbreviation corresponding to the abbreviation target phrase is not stored in the abbreviation dictionary 202. In this case, an abbreviation of the abbreviation target phrase is generated by the abbreviation generator 107, but only the abbreviation generator 107 may be used.

  The image input unit 12 may be a separate scanner device connected to the translation device 1 via a network. The image forming unit 6 may be a separate printer device connected to the translation device 1 via a network.

2 is a diagram illustrating a hardware configuration of a translation apparatus 1. FIG. 2 is a diagram illustrating a functional configuration of a translation apparatus 1. FIG. It is a figure which shows the flow of operation | movement of the translation apparatus. This is an example of the original text written in Japanese. It is an example of the translation sentence which translated the original sentence into English. It is a figure which shows the example of the abbreviation dictionary. It is a figure which shows the example of the translation using an abbreviation. It is a figure which shows the example of the table | surface described in Japanese. It is a figure which shows the example of the table | surface created using the abbreviation.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 ... Translation apparatus, 4 ... Control part, 5 ... Memory | storage part, 41 ... Instruction input part, 9 ... Paper feed tray, 10 ... Sheet, 12 ... Image input part, 6 ... Image formation part, 7Y, 7M, 7C, 7K DESCRIPTION OF SYMBOLS ... Image formation engine, 8 ... Transfer belt, 201 ... Dictionary, 202 ... Abbreviation dictionary, 44 ... CPU, 45 ... ROM, 46 ... RAM, 101 ... Input means, 102 ... Image analysis means, 103 ... Character recognition means, 104 ... translation means, 105 ... determination means, 106 ... extraction means, 107 ... abbreviation generation means, 108 ... abbreviation determination means, 109 ... conversion means, 110 ... image generation means, 111 ... output means.

Claims (18)

A dictionary in which words of the same meaning are stored in association with each other in different languages,
An input means for inputting an image of a document in which the original document is described;
Image analysis means for analyzing a layout of a document represented by an image input by the input means and recognizing a unit area constituting the document;
Character recognition means for recognizing the original text string contained in the image input by the input means;
Translation means for extracting a translation word corresponding to the character string recognized by the character recognition means from the dictionary and generating a translation sentence;
For each unit area recognized by the image analysis means, it is determined whether or not the translated sentence corresponding to the original sentence arranged in the unit area can be arranged in the unit area in the same layout as the original sentence. Determination means to perform,
As a result of the determination by the determination means, if the translated sentence cannot be arranged in the unit area in the same layout as the original sentence, the phrase that satisfies a first condition that is determined in advance is used as the abbreviation target phrase Extraction means for extracting from the sentence;
Abbreviation determination means for determining an abbreviation obtained by omitting a part of the character string constituting the abbreviation target phrase extracted by the extraction means;
Conversion means for generating a translated sentence obtained by converting the abbreviation target phrase extracted by the extraction means into the abbreviation determined by the abbreviation determination means;
An image generation unit that generates an image of the translation sentence generated by the conversion unit in the same layout as the original sentence, and generates an image representing the definition of the abbreviation determined by the abbreviation determination unit;
A translation apparatus comprising: output means for outputting the image generated by the image generation means.   When the translated text generated by the converting means is determined using the determining means, and as a result of the determination, the translated text cannot be arranged in the unit area in the same layout as the original text And abbreviation re-determination means for extracting a phrase satisfying a predetermined second condition from the translation as an abbreviation target phrase and supplying the abbreviation target phrase to the abbreviation determination means. Item 2. The translation device according to Item 1. An abbreviation dictionary storing abbreviations in which a part of a character string constituting a phrase is omitted in association with the phrase;
The translation device according to claim 1, wherein the abbreviation determination unit extracts an abbreviation corresponding to the abbreviation target phrase extracted by the extraction unit from the abbreviation dictionary. , Abbreviation generation means for generating abbreviations in which a part of a character string constituting a phrase is omitted according to a predetermined rule;
When the abbreviation corresponding to the abbreviation target phrase extracted by the extraction means is not stored in the abbreviation dictionary, the abbreviation determination means determines the abbreviation of the abbreviation target phrase by the abbreviation generation means. The translation device according to claim 3, wherein the translation device is generated. Abbreviation generation means for generating abbreviations in which a part of a character string constituting a phrase is omitted according to a predetermined rule;
The translation device according to claim 1, wherein the abbreviation determination means generates an abbreviation corresponding to the abbreviation target phrase extracted by the extraction means by the abbreviation generation means.   5. The translation apparatus according to claim 4, further comprising abbreviation storage means for storing the abbreviation generated by the abbreviation generation means in the abbreviation dictionary in association with the abbreviation target phrase.   When the abbreviation generated by the abbreviation generator is associated with another abbreviation target phrase and is already stored in the abbreviation dictionary, abbreviation reproduction for generating an abbreviation different from the abbreviation The translation apparatus according to claim 4, further comprising a generating unit.   The translation apparatus according to claim 1, wherein the extracting unit extracts a noun or a noun phrase having a number of characters exceeding a predetermined upper limit value as an abbreviation target phrase.   The extraction means extracts the noun or noun phrase as an abbreviation target phrase when the number of appearances of the noun or noun phrase exceeds a predetermined upper limit. The translation apparatus according to any one of 1 to 7.   The translation device according to claim 1, wherein the extraction unit extracts the idiom as a subject to be omitted in preference to the word.   The translation device according to claim 1, wherein the extraction unit extracts an abbreviation target phrase so that the number of abbreviations used in the translated sentence is minimized.   The abbreviation generation means generates an abbreviation using a predetermined number of characters from the head of the abbreviation target phrase when the abbreviation target phrase is a word. The translation device described in 1.   The abbreviation generation means generates an abbreviation by arranging a leading character of the abbreviation target phrase and a consonant following the leading character in the order of appearance when the abbreviation target phrase is a word. The translation device according to claim 4.   The translation apparatus according to claim 13, wherein the abbreviation generation unit uses only a consonant that appears first among the consonants.   The abbreviation generation means generates an abbreviation by arranging the first characters of each word constituting the abbreviation target phrase in the order of appearance when the abbreviation target phrase is a idiom. 8. The translation device according to any one of items 7.   When the abbreviation target phrase is a idiom, the abbreviation generation means configures the first character of each word excluding the word that appears last among the words constituting the abbreviation phrase and the word that appears last The translation device according to claim 4, wherein abbreviations are generated by arranging all characters in the order of appearance. An input step for inputting an image of the original document with the original text;
An image analysis step of analyzing a layout of the document represented by the image input in the input step and recognizing a unit area constituting the document;
A character recognition step for recognizing the original text string included in the image input in the input step;
A translation step of generating a translated sentence by extracting a translation word corresponding to the character string recognized in the character recognition step from a dictionary in which words having the same meaning are associated and stored between different languages;
For each unit area recognized in the image analysis step, it is determined whether a translated sentence corresponding to the original sentence arranged in the unit area can be arranged in the unit area in the same layout as the original sentence A determination step to:
As a result of the determination in the determination step, when the translated sentence cannot be arranged in the unit area with the same layout as the original sentence, the phrase that satisfies the first condition defined in advance is used as the abbreviation target phrase An extraction step to extract from the sentence;
An abbreviation determination step for determining an abbreviation obtained by omitting a part of the character string constituting the abbreviation target phrase extracted in the extraction step;
A conversion step for generating a translation sentence obtained by converting the abbreviation target phrase extracted in the extraction step into the abbreviation determined in the abbreviation determination step;
An image generation step for generating an image of the translated sentence generated in the conversion step in the same layout as the original sentence, and generating an image representing the definition of the abbreviation determined in the abbreviation determination step;
An output step of outputting the image generated in the image generation step. Computer equipment,
A dictionary in which words of the same meaning are stored in association with each other in different languages,
An input means for inputting an image of a document in which the original document is described;
Image analysis means for analyzing a layout of a document represented by an image input by the input means and recognizing a unit area constituting the document;
Character recognition means for recognizing the original text string contained in the image input by the input means;
Translation means for extracting a translation word corresponding to the character string recognized by the character recognition means from the dictionary and generating a translation sentence;
For each unit area recognized by the image analysis means, it is determined whether or not the translated sentence corresponding to the original sentence arranged in the unit area can be arranged in the unit area in the same layout as the original sentence. Determination means to perform,
As a result of the determination by the determination means, if the translated sentence cannot be arranged in the unit area in the same layout as the original sentence, the phrase that satisfies a first condition that is determined in advance is used as the abbreviation target phrase Extraction means for extracting from the sentence;
Abbreviation determination means for determining an abbreviation obtained by omitting a part of the character string constituting the abbreviation target phrase extracted by the extraction means;
Conversion means for generating a translated sentence obtained by converting the abbreviation target phrase extracted by the extraction means into the abbreviation determined by the abbreviation determination means;
An image generation unit that generates an image of the translation sentence generated by the conversion unit in the same layout as the original sentence, and generates an image representing the definition of the abbreviation determined by the abbreviation determination unit;
A program for causing an image generated by the image generating means to function as an output means.

Images