JP2009080727A - Translation device and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a translation device and a program, improving convenience of a user when simultaneously outputting an original text and a translated text. <P>SOLUTION: This translation device 2 reads a document including the original sentence to acquire an original image. The translation device 2 specifies an original text area included in the document image. The translation device 2 recognizes the original text comprising characters included in the original text area and generates the translated text. The translation device 2 arranges the translated text inside the original text area of the document image as parent characters, and arranges the original text as kana characters alongside it to generate a translated document. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a translation apparatus and a program.

There is a device that analyzes a scanned image of a document including an original text, and translates and outputs the original text included in the scanned image. For example, the following Patent Document 1 describes a method of outputting an image in which an original sentence and a translated sentence are alternately arranged in line units. According to this method, for example, the user can easily confirm the translation result.
JP-A-5-324720

  However, in the method of Patent Document 1, since the character size of the translated sentence is made smaller than the character size of the original sentence, the character size of the translated sentence becomes too small depending on the character size of the original sentence, and it is easy to check the translation result. May disappear. To begin with, translations tend to be more important for users than for originals.

  The present invention has been made in view of the above problems, and one of its purposes is to provide a translation device and a program that improve the convenience of the user when outputting an original sentence and a translated sentence simultaneously. To do.

  According to the first aspect of the present invention, there is provided a document image acquiring unit that acquires a document image including a source region, a source region specifying unit that specifies the source region of the document image, and characters included in the source region of the document image. A translation means for recognizing the original sentence to generate a translation sentence, and a translation document in which the translation sentence is arranged as a parent character in the original sentence area of the manuscript image and the original sentence is arranged as a ruby character to generate a translation document And a generation unit. Here, the ruby character is placed along the right or left of the character string of the translated sentence (parent character string) when the translated sentence is written vertically in the translated document, and the translated sentence is written horizontally. Is a character that is placed above or below the character string of the parent character and whose character size is smaller than the character size of the parent character.

  According to a second aspect of the present invention, in the first aspect of the present invention, the character size specifying means for specifying the character size of the original text included in the original text area, and the character size specified by the character size specifying means for the character size of the parent character. Character size determining means for determining based on the size is further included.

  According to a third aspect of the present invention, in the second aspect of the invention, a translated sentence comprising at least one line based on the size of the original text area, the number of characters of the translated sentence, and the determined character size of the parent character. A translated sentence area setting means for regularly setting a translated sentence area in which the text is arranged in the original sentence area, and a ruby character area in which the source sentence consisting of at least one line is set between the translated sentence areas. And a character area setting means.

  The invention of claim 4 further includes character line image acquisition means for acquiring a character line image of each line of the original text included in the original text area according to the invention of claim 3, wherein the translated document generation means Each character line image is arranged in an area.

  According to a fifth aspect of the present invention, in the invention of the fourth aspect, the judging means for judging whether or not the height of each character line image is larger than the interval between the translated sentence regions, and the height is higher than the interval. It further includes first reduction means for reducing the character line image determined to be large so that the height thereof is equal to or less than the interval.

  The invention of claim 6 is the invention of claim 4 or 5, wherein when the number of character line images is larger than the number of lines of translated sentences arranged in the source text area, the length of each character line image is set. A second reduction unit that reduces the image at a magnification smaller than 1 and divides the connected images of the character line images reduced by the second reduction unit by the length of each ruby character region, A character line image is used.

  According to a seventh aspect of the present invention, in the sixth aspect of the present invention, the magnification smaller than 1 is a magnification equal to or lower than a quotient obtained by dividing the number of translated text lines arranged in the source text area by the number of character line images. It is characterized by that.

  The invention according to claim 8 is a document image acquisition means for acquiring a document image including a text area, a text area specifying means for specifying the text area of the text image, and a text composed of characters included in the text area of the text image. A computer functioning as a translation document generating means for recognizing and translating a text, arranging a translated text as a parent character in the original text area of the manuscript image and arranging the original text as a ruby character to generate a translated document It is a program to let you.

  According to the first and eighth aspects of the invention, it is possible to improve the convenience for the user when the original text and the translated text are output simultaneously.

  According to the invention of claim 2, the character size of the parent character can be determined based on the character size of the original text in the original text image.

  According to the invention of claim 3, the original sentence and the translated sentence can be output regularly.

  According to the invention of claim 4, the original text can be output without error.

  According to the invention of claim 5, it is possible to prevent the ruby character from overlapping the parent character.

  According to the invention of claim 6, the number of lines in the original sentence can be made smaller than the number of lines in the translated sentence.

  According to the invention of claim 7, the number of lines of the original sentence can be surely reduced from the number of lines of the translated sentence.

  Embodiments of the present invention will be described with reference to the drawings. FIG. 1 is an example of a hardware configuration diagram of a translation apparatus 2 according to an embodiment of the present invention. As shown in the figure, the translation apparatus 2 includes a central processing unit 40, a main storage unit 42, an auxiliary storage unit 44, an image reading unit 46, an operation unit 48, and a display 50. Each unit can communicate with each other via a data bus 54. The translation device 2 includes a network interface, a printer, etc. (not shown).

  The central processing unit 40 is a CPU, MPU, or the like. The central processing unit 40 operates according to a program stored in advance in the main storage unit 42 and controls each unit. In addition, the central processing unit 40 calculates using information stored in the main storage unit 42 and outputs it to the main storage unit 42. The program is not limited to the one stored in the main storage unit 42 but may be stored in an information storage medium such as a CD-ROM, DVD-ROM, or DVD-ROM, or supplied from a network. It may be done.

  The main storage unit 42 is a memory element such as a RAM or a ROM. The main storage unit 42 stores the program. Further, the main storage unit 42 stores information input from each unit.

  The auxiliary storage unit 44 is a hard disk or the like. The auxiliary storage unit 44 stores information stored in the main storage unit 42 in accordance with a control signal from the central processing unit 40.

  The image reading unit 46 is a scanner. The image reading unit 46 reads a document in accordance with a control signal from the central processing unit 40 and outputs the read document image to the main storage unit 42. Here, the manuscript includes the original text in the first language. Here, the first language is Japanese.

  The operation unit 48 is a keyboard, a mouse, or the like. The operation unit 48 outputs operation information to the main storage unit 42. The display 50 displays information stored in the main storage unit 42 in accordance with a control signal from the central processing unit 40.

  FIG. 2 is a block diagram of functions realized by the translation apparatus 2 when the central processing unit 40 operates according to the program. Here, the functions related to the present invention are mainly shown. As shown in the figure, the translation apparatus 2 includes a document image acquisition unit 60, a source text region specification unit 62, a character line image acquisition unit 64, a translation unit 66, a parent character size determination unit 68, and an arrangement region setting. A section 70, a translation document generation section 76, and a translation result output section 78. The arrangement region setting unit 70 includes a parent character region setting unit 72 and a ruby character region setting unit 74.

  The document image acquisition unit 60 is realized with the image reading unit 46 as the center. The document image acquisition unit 60 reads a document and acquires image data (bitmap data) of the document image.

  The original text region specifying unit 62 is realized with the central processing unit 40 as the center. The original text area specifying unit 62 analyzes the layout of the original image acquired by the original image acquisition unit 60 and specifies the position / range of the original text area included in the original image. In the present embodiment, as shown in FIG. 3, the original text area specifying unit 62 detects the original text area as a rectangular area surrounding the original text image, and specifies the range of the X and Y coordinates.

  The character line image acquisition unit 64 is realized with the central processing unit 40 as the center. The character line image acquisition unit 64 acquires, for each original text area, a character line image of each line of the original text included in the original text area.

  In the present embodiment, as shown in FIG. 4, the character line image acquisition unit 64 performs a projection operation for each original text region, and generates a projection histogram of character pixels with respect to the vertical axis (Y axis). Then, the character line image acquisition unit 64 specifies the position / range of the character line region for each original text region based on the projection histogram. That is, the character line image acquisition unit 64 specifies a range of Y coordinates in which the number of character pixels is equal to or greater than a predetermined value TH, and sets the specified range as a range in the vertical direction of each character line region. Further, the range of the X coordinate of the original text area is set as the horizontal range of each character line area. Thus, the character line image acquisition unit 64 specifies a character line area for each original text area as shown in FIG.

  And the character line image acquisition part 64 duplicates the image (namely, character line image) in each character line area | region for every original text area | region. Thus, the character line image acquisition unit 64 acquires each character line image for each original text area.

  The translation unit 66 is realized with the central processing unit 40 as the center. The translation unit 66 recognizes an original sentence composed of characters included in the original sentence area and generates a translated sentence in the second language.

  In the present embodiment, the translation unit 66 recognizes the original text included in the original text area by recognizing the characters included in each character line image acquired from the original text area for each original text area. . And the translation part 66 produces | generates the text of the translated sentence of the recognized original sentence for every original sentence area | region. Here, the second language is English.

  The parent character size determination unit 68 is realized with the central processing unit 40 as the center. The parent character size determination unit 68 determines, for each source text area, a character size when a translation document generation unit 76, which will be described later, arranges the text of the translation sentence as a parent character in the source text area. Here, the character size is the font size.

  In the present embodiment, the parent character size determination unit 68 specifies the character size of the original text included in the original text area from the vertical width of the character line image acquired from the original text area for each original text area, and specifies the specified character. A character size larger than the size (for example, a character size one point larger than the specified character size) is set as the character size of the parent character arranged in the original text area.

  The arrangement area setting unit 70 is realized with the central processing unit 40 as the center. The arrangement region setting unit 70 includes a parent character region setting unit 72 and a ruby character region setting unit 74.

  The parent character area setting unit 72 sets, for each source text area, a parent character that arranges a parent character based on the size of the source text area, the number of characters in the translated sentence, and the character size of the parent text placed in the source text area. Set the character area in the source text area.

  In the present embodiment, the parent character area setting unit 72 uses a line of parent characters that can be placed in the source text area based on the width of the source text area and the width of the character size of the parent text placed in the source text area. Calculate the number of characters per. Then, the parent character area setting unit 72 calculates the number of lines when the parent characters are arranged in the original sentence area from the number of characters of the translated sentence (number of characters of the parent character). For example, if the number of characters per line that can be arranged in the original text area is 100 characters and the number of characters of the parent character is 450 characters, the calculated number of lines is 5.

  The parent character area setting unit 72 sets the same number of arrangement areas as the calculated number of lines in the original sentence area. Specifically, the parent character area setting unit 72 sets each rectangular area formed by equally dividing the original text area by the calculated number of lines as an arrangement area. FIG. 6A shows an example of an arrangement area set in the original sentence area when the calculated number of lines is three.

  Then, as shown in FIG. 6B, the parent character area setting unit 72 sets the vertical width of the character size of the parent character as the vertical width and the horizontal width of the original text area as the horizontal width at the bottom of each arrangement area. Set the parent character area. Therefore, each parent character area is set at equal intervals.

  The ruby character area setting unit 74 sets, for each original text area, a ruby character area for arranging original text characters as ruby characters in the original text area. In the present embodiment, the ruby character area setting unit 74 sets one ruby character area in each arrangement area. Specifically, as shown in FIG. 6B, the ruby character area setting unit 74 sets a rectangular area other than the parent character area in the arrangement area as the ruby character area. Accordingly, the vertical width of each ruby character area is a difference obtained by subtracting the vertical width of the parent character area from the vertical width of the arrangement area, and the horizontal width of each ruby character area is the horizontal width of the original text area.

  The translated document generation unit 76 arranges the translated text as a parent character in each original text area of the original image, and arranges the original text as a ruby character to generate a translated document. FIG. 7 shows a translated document generated based on the document image shown in FIG.

  In the present embodiment, the translation document generation unit 76 fills each original text area of the document image with a specific color. Then, as shown in FIG. 8, the translated document generation unit 76 arranges each character line image acquired from the original text area in each ruby character area set in the original area for each original area.

  However, the translation document generation unit 76 arranges the character line images after reducing them in some cases.

  That is, when the translated document generation unit 76 arranges a character line image acquired from the original text region in each ruby character region in the original text region of interest, the translated document generation unit 76 selects from the character line images acquired from the original text region. The one with the largest vertical width (attention character line image) is selected, and it is determined whether or not the vertical width of the attention character line image is larger than the vertical width of the ruby character area. If the vertical width of the target character line image is larger than the vertical width of the ruby character area, the vertical width of each character line image is reduced as shown in FIG. Here, the vertical width of each character line image is reduced at a magnification equal to or lower than the quotient obtained by dividing the vertical width of the ruby character area by the vertical width of the target character line image. As a result, the vertical width of each character line image is reduced below the vertical width of the ruby character area.

  In addition, when the translated document generation unit 76 places a character line image acquired from the original text area in each ruby character area in the original text area of interest, the number of parent character areas set in the original text area If the number of character line images acquired from the original text area is larger than the width of each character line image, as shown in FIG. For example, the translation document generation unit 76 reduces the width of each character line image at a magnification equal to or lower than the quotient obtained by dividing the number of parent character regions by the number of character line images. The magnification may be a predetermined magnification smaller than 1.

  Then, the translation document generation unit 76 converts each character line image reduced in at least one of the vertical width and the horizontal width as described above into a ruby character area, and the translation document generation unit 76 converts each character line image When the horizontal width of the image is reduced, as shown in FIG. 11A, the character line images are connected in the horizontal direction and divided by the horizontal width of the ruby character area. Then, as shown in FIG. 11B, each divided image is arranged in each ruby character area as a new character line image.

  When each character line image is arranged in each ruby character area as described above, the translation document generation unit 76 uses the character size determined by the parent character size determination unit 68 in each parent character area, and the character image of the translated sentence. Are placed as parent characters. Thus, the translation document generation unit 76 generates image data (bitmap data) of the translation document.

  The translation result output unit 78 is realized with the central processing unit 40 as the center. The translation result output unit 78 causes the display 50 to output the translation document generated by the translation document generation unit 76. The translation result output unit 78 may cause the printer to print out the translated document.

  Next, an example of processing executed by the translation apparatus 2 having the above function will be described with reference to the flowchart of FIG. In the present embodiment, this process is executed when the user causes the translation apparatus 2 to read a document and performs a translation instruction operation.

  The translation device 2 reads the document and acquires the image data of the document image by the function of the document image acquisition unit 60. Then, the translation apparatus 2 performs known image quality enhancement processing such as skew correction and background removal processing on the acquired document image (S101). Then, the translation device 2 analyzes the layout of the document image by using the function of the original text region specifying unit 62 and specifies the position / range of the original text region included in the original image (S102).

  Then, the translation device 2 specifies the character line area of each line of the original sentence for each original text area by the function of the character line image acquisition unit 64, and duplicates the image (that is, the character line image) in each character line area. (S103).

  And the translation apparatus 2 recognizes the character of the original text contained in the said character line image for every character line image acquired by the step of S103 by a well-known process (S104). And the translation apparatus 2 recognizes the original text contained in the said original text area | region for every original text area | region by integrating the character recognized for every character line image for every original text area | region (S105). And the translation apparatus 2 produces | generates the text of the recognized translation of the original sentence for every original text area | region (S106).

  Then, the translation apparatus 2 selects one target original text area from the original text areas specified in step S102 (S107). Then, the translation apparatus 2 fills the target original text area with a predetermined color (for example, white) (S108). Here, the translation apparatus 2 may detect an average color of an area other than the character line area in the target original text area, and fill the target text area with the average color.

  Then, the translation apparatus 2 determines the character size of the parent character to be arranged in the target original text area by the function of the parent character size determination unit 68 (S109).

  That is, the translation device 2 focuses on one of the character line images acquired from the target original text area, and specifies the character size of the original text included in the target text area based on the vertical width. Then, the translation apparatus 2 determines a character size that is one point larger than the specified character size as the character size of the parent character to be arranged in the target original text area.

  Further, the translation device 2 detects the character color of the original text included in the target original text region based on the focused character line image, and determines a character color different from the detected character color as the character color of the parent character. (S110). For example, the translation device 2 determines a character color having a predetermined color difference from the detected character color as the character color of the parent character.

  Then, the translation device 2 calculates the number of lines when the parent character is arranged in the target text area by the function of the parent text area setting unit 72, and sets the same number of arrangement areas as the calculated number of lines in the target text area. Set. Then, translation apparatus 2 sets a parent character area at the bottom of each arrangement area (see FIG. 6). Also, the translation device 2 sets a rectangular area other than the parent character area in each arrangement area as the ruby character area by the function of the ruby character area setting unit 74 (S111).

  Then, the translation apparatus 2 determines whether or not the number of character line images acquired from the target text area is larger than the number of parent character areas set in the target text area (S112). If the number of parent character areas is equal to or greater than the number of character line images (N in S112), the translation device 2 determines that the character line image having the largest vertical width (from the character line images acquired from the original text area of interest) The target character line image) is selected, and it is determined whether or not the vertical width is larger than the vertical width of the ruby character area (S113). If the vertical width of the target character line image is equal to or smaller than the vertical width of the ruby character area (N in S113), the translation device 2 proceeds to S114. On the other hand, if the vertical width of the target character line image is larger than the vertical width of the ruby character area (Y in S113), the translation apparatus 2 sets the vertical width of each character line image to the vertical width of the ruby character area. The image is reduced at a magnification equal to or lower than the quotient divided by the vertical width of the image (S115), and the process proceeds to step S114.

  On the other hand, when the number of character line images acquired from the target text area is larger than the number of parent character areas set in the target text area (Y in S112), the translation apparatus 2 determines the width of each character line image as the parent text area. Reduction is performed at a magnification equal to or less than the quotient obtained by dividing the number of character areas by the number of character line images (S116). Note that the magnification here may be a predetermined magnification smaller than 1. Then, the translation device 2 determines whether or not the vertical width of the target character line image is larger than the vertical width of the ruby character area (S117). If the vertical width of the target character line image is equal to or smaller than the vertical width of the ruby character area (N in S117), the translation apparatus 2 proceeds to S118. On the other hand, if the vertical width of the target character line image is larger than the vertical width of the ruby character area (Y in S117), the translation apparatus 2 sets the vertical width of each character line image to the vertical width of the ruby character area. The image is reduced at a magnification equal to or smaller than the quotient divided by the vertical width of the image (S119), and the process proceeds to step S118. Note that the translation apparatus 2 may reduce each character line image as a whole in steps S116 and S119.

  Then, in step S118, translation apparatus 2 concatenates each character line image in the horizontal direction and divides it by the horizontal width of the ruby character region, and sets each divided image as a new character line image, step S114. Proceed to

  Then, in step S114, the translation device 2 arranges each character line image in each ruby character area, and in each parent character area, a character image of the translated sentence, that is, a parent character image, is placed in step S109. The character size determined in step S120 is arranged (S120).

  Then, the translation apparatus 2 determines whether or not steps S108 to S120 have been executed for all the original text regions (S121). If the steps S108 to S120 are not executed for all the original text regions (N in S121), the translation apparatus 2 selects another target text region and executes the steps after S108. On the other hand, when the steps S108 to S120 are executed for all the original text regions (Y in S121), the translation apparatus 2 displays the translation document based on the image data of the translation document generated by the steps so far. 50 or print out (S122). The above is the specific content of the processing executed by the translation apparatus 2. Note that the translation apparatus 2 may perform the process of FIG. 12 on the original image of each page when the original consists of a plurality of pages.

  As described above, according to the translation device 2 according to the embodiment of the present invention, a translated document in which the original text characters are arranged as ruby characters and the translated text characters are arranged as parent characters is output. Convenience for the user is improved.

  The embodiment of the present invention is not limited to the above embodiment.

  For example, the translation apparatus 2 may generate image data of the translation document by drawing an image of the original text recognized in step S105 in each ruby character area. However, the translation device 2 draws the image of the ruby character (original character) so that the size of the ruby character is smaller than that of the parent character (translated character).

  Further, for example, the translation apparatus 2 may set each parent character area not at the bottom of each arrangement area but at the top in the step of S111. That is, the character string of each line of ruby characters may be arranged below the character string of each line of the parent character.

  For example, the translation apparatus 2 may reduce each character line image as a whole in step S115. In this case, after the step of S115, as in the step of S118, the translation apparatus 2 concatenates each character line image in the horizontal direction and divides it by the horizontal width of the ruby character region, and each divided image is newly added. As a simple character line image, the process proceeds to step S114.

  For example, the translation apparatus 2 may set each arrangement area so that the vertical width gradually increases from the upper part to the lower part of the original sentence area in step S111. Thereby, each parent character area is set so that the interval gradually increases from the top to the bottom of the original text area. In this case, the translation apparatus 2 determines whether or not the vertical width of the target character line image is larger than the vertical width of the ruby character area set in the uppermost arrangement area in steps S113 and S117. In step S115 or S119, it is conceivable to reduce the vertical width of each character line image by a magnification equal to or less than the quotient obtained by dividing the vertical width of the ruby character area by the vertical width of the target character line image.

  Further, for example, the translation device 2 arranges the parent character and the ruby character in the document image itself, but may arrange the parent character and the ruby character in a duplicate image of the document image.

  Moreover, although the translation apparatus 2 has produced | generated the translation document in the image data format in the said embodiment, you may make it produce | generate in another data format. For example, the translation device 2 uses drawing data including data such as a drawing position, a drawing size, and a drawing color of a parent character for a document image (more precisely, a document image in which a character line image is arranged in each original text area). A translation document may be generated.

  In addition, the present invention is applicable regardless of the language of the first language and the second language. For example, the present invention can be applied to a case where French is translated into German.

  Further, the present invention can be applied not only to a document including a horizontally written original but also to a document including a vertically written original.

It is a figure which shows an example of the hardware constitutions of the translation apparatus which concerns on embodiment of this invention. It is a figure which shows an example of the functional block of the translation apparatus which concerns on embodiment of this invention. It is a figure which shows an example of a manuscript image. It is a figure for demonstrating a mode that a character line area is specified. It is a figure for demonstrating a mode that a character line area is reached. It is a figure for demonstrating a mode that an arrangement | positioning area | region, a parent character area | region, and a ruby character area | region are set to an original sentence area | region. It is a figure which shows an example of a translation document. It is a figure for demonstrating a mode that a character line image is arrange | positioned in an original text area | region. It is a figure for demonstrating a mode that the vertical width of a character line image is reduced. It is a figure for demonstrating a mode that the horizontal width of a character line image is reduced. It is a figure for demonstrating a mode that a new character line image is produced | generated. It is a flowchart figure which shows an example of the process which the translation apparatus concerning embodiment of this invention performs.

Explanation of symbols

  2 translation device, 40 central processing unit, 42 main storage unit, 44 auxiliary storage unit, 46 image reading unit, 48 operation unit, 50 display, 54 data bus, 60 original image acquisition unit, 62 original text region specifying unit, 64 character line Image acquisition unit, 66 translation unit, 68 parent character size determination unit, 70 arrangement region setting unit, 72 parent character region setting unit, 74 ruby character region setting unit, 76 translation document generation unit, 78 translation result output unit

Claims (8)

Document image acquisition means for acquiring a document image including a source text area;
Original text area specifying means for specifying the original text area of the original image;
Translation means for recognizing an original text composed of characters included in the original text area of the original image and generating a translated text;
In the original text area of the original image, a translated text is arranged as a parent character, a translated text is arranged as a ruby character, and a translated document is generated to generate a translated document
The translation apparatus characterized by including. Character size specifying means for specifying the character size of the original text included in the original text area;
Character size determining means for determining a character size of a parent character based on the character size specified by the character size specifying means;
The translation apparatus according to claim 1, further comprising: Based on the size of the original text area, the number of characters of the translated text, and the determined character size of the parent character, a translated text area in which the translated text consisting of at least one line is arranged in the original text area. A translation area setting means to be set to
Ruby character area setting means for setting a ruby character area in which an original sentence consisting of at least one line is arranged between the translated sentence areas;
The translation apparatus according to claim 2, further comprising: A character line image acquisition means for acquiring a character line image of each line of the original text included in the original text area;
The translated document generation means arranges each character line image in each ruby character region.
The translation apparatus according to claim 3. Determining means for determining whether or not the height of each character line image is larger than the interval between translated sentence regions;
First reduction means for reducing a character line image whose height is determined to be larger than the interval so that the height is equal to or less than the interval;
The translation apparatus according to claim 4, further comprising: A second reduction means for reducing the length of each character line image at a magnification smaller than 1 when the number of the character line images is larger than the number of lines of translated sentences arranged in the original sentence region;
Dividing the concatenated image of each character line image reduced by the second reduction means by the length of each ruby character region, and making each divided image a new character line image;
The translation apparatus according to claim 4 or 5, characterized in that: The magnification smaller than 1 is a magnification equal to or lower than the quotient obtained by dividing the number of translated text lines arranged in the original text area by the number of character line images.
The translation apparatus according to claim 6. A document image acquisition means for acquiring a document image including an original area;
Original text area specifying means for specifying the original text area of the document image;
Translation means for recognizing and translating an original text consisting of characters included in the original text area of the original image;
A translation document generating means for arranging a translation sentence as a parent character in the original sentence area of the original image and arranging the original sentence as a ruby character to generate a translation document;
As a program that allows the computer to function.

Images