JP2004362073A - Document processor and method and control program for document processor - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a document processor and method enabling a user to efficiently understand the contents of documents, while keeping the portion of summary documents at an appropriate amount; and to provide a control program for the document processor. <P>SOLUTION: The document processor comprises a component element extracting part 102 for extracting at least text elements and image elements from documents; a text summarizing process part 105 for summarizing the text elements; and a summary document composing part 107 for composing summary documents from the summarized texts generated by the text summarizing process part 105 and the image elements selected by an image selecting part 106. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a document processing apparatus and a document processing method for summarizing a document, and a control program for the document processing apparatus.
[0002]
[Prior art]
With the widespread use of the Internet and PCs (personal computers), in recent years, an enormous amount of documents have been created and distributed. Since the amount of generated documents exceeds the processing capacity of the user, it is becoming difficult for the user to obtain necessary information from a large number of documents.
[0003]
As one solution to such a problem, a technique for summarizing text as described in Non-Patent Document 1 has been developed. With this technique, the text in the document can be summarized and presented, so that the user can efficiently grasp the contents of the text.
[0004]
On the other hand, as another solution, Patent Literature 1 discloses a method in which each page of a document is reduced and printed so that it can be viewed, and a user can easily determine the necessity of the document.
[0005]
In Patent Document 2, a document is classified into a structured document and a document image. In the case of a structured document, a text summary is created and presented using logical structure information of the text. Discloses a method of presenting individual components in a reduced size. Here, the document image refers to an image in which document data including a text portion is represented as an image, such as an image obtained by scanning a paper document.
[0006]
[Non-patent document 1]
Okumura Manabu, Namba Eiji, "Research Trend on Automatic Text Summarization", Natural Language Processing, Special Issue on "Linguistic Processing for Text Summarization", Vol. 6, no. 6, July 1999 [Patent Document 1]
JP-A-7-146875 [Patent Document 2]
JP-A-10-320412
[Problems to be solved by the invention]
In general, most documents include not only text elements but also image elements such as charts, photographs, and images. When a summary document is created for such a document, the text summarization method results in the image element remaining even though the text is summarized.
[0008]
When a large number of image elements exist in a document, the amount of the entire summary document increases when the document is output on paper or the like, so that it is difficult for a user to easily determine the contents of the document.
[0009]
Also, as described in Patent Document 1 and Patent Document 2 described above, in the method of printing each page in a reduced size, the amount of the summary document can be reduced. Is small, it is difficult for the user to understand the contents of the text.
[0010]
SUMMARY OF THE INVENTION The present invention has been made to solve the above-described problems of the related art, and an object of the present invention is to provide a document that allows a user to efficiently grasp the contents of a document while suppressing the amount of a summary document to an appropriate amount. An object of the present invention is to provide a processing device, a document processing method, and a control program for the document processing device.
[0011]
[Means for Solving the Problems]
In order to achieve the above object, a document processing apparatus according to the present invention comprises: a component extracting unit configured to extract at least a text element and an image element from a document; a summary text generating unit configured to generate a summary text from the text element; A summary image generating means for generating a summary image from an element, and a summary document generating means for generating a summary document from the summary text and the summary image are provided.
[0012]
Further, in order to achieve the above object, the document processing method of the present invention includes a component extraction step of extracting at least a text element and an image element from a document; a summary text generation step of generating a summary text from the text element; A summary image generating step of generating a summary image from the image elements; and a summary document generating step of generating a summary document from the summary text and the summary image.
[0013]
Further, in order to achieve the above object, a control program for a document processing apparatus according to the present invention is characterized by comprising a program code for causing a computer to execute each step of the document processing method.
[0014]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, embodiments of the present invention will be described with reference to the drawings.
[0015]
(First Embodiment)
First, a first embodiment of the present invention will be described with reference to FIGS.
[0016]
FIG. 1 is a block diagram showing a basic configuration of a document processing apparatus according to the present embodiment. In FIG. 1, reference numeral 101 denotes a document holding unit which holds a document to be processed. A component extraction unit 102 extracts a text element and an image element from the document held in the document holding unit 101. In the present embodiment, it is assumed that the document is a digitized document having a structure, and that components such as texts, charts, and images can be easily obtained. Note that elements other than text are treated as image elements here.
[0017]
A text element holding unit 103 holds the text elements extracted by the component extraction unit 102. An image element holding unit 104 holds the image elements extracted by the component extracting unit 102. A text summarizing unit 105 summarizes the text elements held in the text element holding unit 103. An image selection unit 106 selects some of the image elements held in the image element holding unit 104. Reference numeral 107 denotes a summary document synthesis unit that synthesizes a summary document from the summary text generated by the text summary processing unit 105 and the image element selected by the image selection unit 106. Reference numeral 108 denotes a summary document holding unit which holds the summary document synthesized by the summary document synthesis unit 107. A summary document output unit 109 outputs the summary document held in the summary document holding unit 108.
[0018]
FIG. 2 is a block diagram showing a specific configuration of the document processing apparatus according to the present embodiment. In FIG. 2, reference numeral 201 denotes a CPU (central processing unit), which operates according to a program for implementing a procedure described later. Reference numeral 202 denotes a memory that implements the document holding unit 101, the text element holding unit 103, the image element holding unit 104, and the summary document holding unit 108, and provides a storage area necessary for the operation of the program. Reference numeral 203 denotes a control memory which holds a program for implementing a procedure described later. An output device 204 implements the summary document output unit 109. Specifically, it is a display device or a printer device. A bus 205 connects the components.
[0019]
Next, the operation of the document processing apparatus according to the present embodiment will be described with reference to the flowchart in FIG.
[0020]
First, in step S301, the component extracting unit 102 extracts a document component from the document held in the document holding unit 101. The document component is a text element or an image element. In the present embodiment, as described above, the document is a digitized document having a structure, and these elements can be easily extracted. The text element is held in the text element holding unit 103, and the image element is held in the image element holding unit 104.
[0021]
FIG. 6 shows an example of the contents of the text element holding unit 103. An ID (identifier) is assigned to each text constituent unit, and holds a text character string and an ID of a referenced image element. Such a reference relationship is described in the electronic document and can be easily extracted. Here, the constituent unit of the text is a sentence here, but another constituent unit may be adopted.
[0022]
FIG. 7 shows an example of the contents of the image element holding unit 104. An ID is assigned to each image element, and holds a pointer to image data and an ID of a text element referencing the image element.
[0023]
Referring back to FIG. 3, in step S302, it is checked whether or not there is a text element in the document components extracted in step S301. If there is a text element, the process proceeds to step S303. If there is no text element, the process proceeds to step S308.
[0024]
In step S303, the text summarization processing unit 105 performs a summarization process on the text elements held in the text element holding unit 103. Here, an abstract is generated by selecting an important sentence from the text. Specifically, it can be performed by a generally known text summarization method as described in Non-Patent Document 1, and an arbitrary amount of summary text can be generated. The amount of summary text to be generated is given by a numerical value (the number of characters or the ratio to the number of characters in the original document). This numerical value may be set in advance, may be given by the user at the time of processing, or may be set automatically. As a result of the summarization, the text element holding unit 103 adds a mark to the selected sentence as shown in FIG.
[0025]
Next, in step S304, it is checked whether or not there is an image element in the document components extracted in step S301. If there is an image element, the process proceeds to step S305, and if there is no image element, the process proceeds to step S306.
[0026]
In step S305, the image selecting unit 106 selects, from among the image elements held in the image element holding unit 104, those related to the summary text generated in step S303. As shown in FIG. 7, the image element holding unit 104 marks the image element selected in step S305 in the summary column.
[0027]
The selection of the image element in step S305 is specifically performed by selecting an image element referred to from the summary text. This reference relationship is held in the image element holding unit 104. In the examples of FIGS. 6 and 7, the sentence of ID = tx1 is included in the summary text, but the sentences of ID = tx2 and tx3 are not included. In this case, the image element im1 referenced from the sentence with ID = tx1 is selected, and the image element im2 referenced from the sentence with ID = tx2 is not selected.
[0028]
Next, in step S306, the summary document synthesizing unit 107 extracts the summary document from the summary text held in the text element holding unit 103 and the selected one of the image elements held in the image element holding unit 104. Generate. When the text element holding unit 103 does not hold the summary text, the summary document is generated only from the image element, and when the image element holding unit 104 does not hold the image element, the summary document is generated only from the summary text. . The generated summary document is stored in the summary document storage unit 108.
[0029]
Next, in step S307, the summary document held in the summary document holding unit 108 is output to the summary document output unit 109, and then this processing operation ends.
[0030]
On the other hand, if the text element does not exist in the text element holding unit 103 in step S302, the process proceeds to step S308, and it is determined whether or not the image element is held in the image element holding unit 104. If the image element is held, the process proceeds to step S309. If the image element is not held, the process ends.
[0031]
In step S309, a part of the image elements held in the image element holding unit 104 is selected, and the process proceeds to step S306. There is no particular limitation on the selection method. The amount of the image element selected here (the number of elements or the ratio to the total number of elements) may be given by a numerical value. This numerical value may be set in advance, or may be given by the user at the time of processing. Among the image elements held in the image element holding unit 104, a mark is given to the one selected in step S309.
[0032]
Next, a more detailed description will be given based on a specific example with reference to FIGS.
[0033]
In the example shown in FIG. 4, an image element 402 is referenced from a character string 401 in a text element, and an image element 405 is referenced from a character string 404 in a text element. Here, it is assumed that the generated summary text includes the character string 401 and the character string 407 but does not include the character string 404. In this case, the image element 402 referenced from the character string 401 is selected, and the image element 405 referenced from the character string 404 is not selected. In the case of this document, an output as shown in FIG. 5 is obtained as a summary document. The character strings 401 and 407 included in the abstract text and the image element 402 referred to therefrom are output, and the character string 404 not included in the abstract text and the image element 405 referenced therefrom are not output. Thus, only the summary text and the necessary image elements are output to the summary document. In this example, the caption is output for the image element, but may not be output.
[0034]
(Second embodiment)
In the first embodiment, when selecting an image element related to the summary text in step S305 in FIG. 3, the image element referred to from the summary text is selected. May be selected. Here, the constitutional unit refers to a chapter, a section, a section, a paragraph, and the like, and which constitutional unit is to be based on is set in advance. For example, image elements included in a paragraph containing the summary text are selected, and image elements that are not are not selected.
[0035]
(Third embodiment)
In the first embodiment, when selecting an image element related to the text summary in step S305 in FIG. 3, the image element referred to in the summary text is selected. If added, an image element having a caption similar to the summary text may be selected. To determine the similarity between the summary text and the caption, for example, a generally known vector space model (for example, described in “Salton, McGill,“ An Introduction to Modern Information Retrieval ”, 1983.”) is used. Both the summary text and the caption can be expressed as vectors, and the similarity can be determined based on the distance between the vectors. However, another method may be used.
[0036]
When determining the similarity, not all of the summary texts may be used, but some of them may be used. For example, it is conceivable to use a summary text existing near an image element.
[0037]
(Fourth embodiment)
In the first embodiment, when selecting an image element related to the text summary in step S305 in FIG. 3, the image element referred to from the summary text is selected. In addition, a user may explicitly specify an image element to be output to the summary document.
[0038]
(Fifth embodiment)
In the first embodiment, a part of the image elements is selected and output in the summary document. However, instead of selecting the image elements, all the image elements are uniformly reduced to be included in the summary document. You may make it output. In this case, image reduction processing means is provided in place of the image selection means, and processing for reducing all image elements is performed instead of image element selection processing. Here, the reduction ratio may be given by a numerical value. This numerical value may be set in advance, or may be given by the user at the time of processing.
[0039]
Further, the selection of the image element and the reduction of the image element may be used together.
[0040]
Further, an image element output method setting unit may be provided, and the user may specify whether to select and output the image element, uniformly reduce the image element and output the image element, or use both of them. .
[0041]
(Sixth embodiment)
Further, in the first embodiment, a digitized document having a structure is processed, but the present invention can also be implemented for a document image obtained by scanning.
[0042]
Further, by providing an area separating means for extracting a text area and an image area from a document image and a character recognizing means for extracting a text character string from the text area thus extracted, an image element and a text element can be obtained. , Can be carried out in the same manner as in the first embodiment.
[0043]
(Seventh embodiment)
Further, in the first embodiment, the case where each unit is configured on the same electronic computer has been described. However, the present invention is not limited to this, and may be realized on a plurality of electronic computers.
[0044]
(Other embodiments)
Note that the present invention may be applied to a system including a plurality of devices or to an apparatus including a single device.
[0045]
In addition, a storage medium storing software program codes for realizing the functions of the above-described embodiments is supplied to a system or apparatus, and a computer (or CPU or MPU) of the system or apparatus stores the program stored in the storage medium. Needless to say, this can also be achieved by reading and executing the code.
[0046]
In this case, the program code itself read from the storage medium realizes the functions of the above-described embodiment, and the storage medium storing the program code constitutes the present invention.
[0047]
Examples of a storage medium for supplying the program code include a floppy (registered trademark) disk, a hard disk, an optical disk, a CD-ROM, a CD-R, a DVD-ROM, a magnetic tape, a nonvolatile memory card, and a ROM. Can be used.
[0048]
When the computer executes the readout program codes, not only the functions of the above-described embodiments are realized, but also an OS (Operating System) or the like running on the computer based on the instructions of the program codes. Does part of or all of the actual processing, and the processing realizes the functions of the above-described embodiments.
[0049]
Further, after the program code read from the storage medium is written into a memory provided in a function expansion board inserted into the computer or a function expansion unit connected to the computer, the function expansion is performed based on the instruction of the program code. It goes without saying that the CPU or the like provided in the board or the function expansion unit performs part or all of the actual processing, and the processing realizes the functions of the above-described embodiments.
[0050]
Although various examples and embodiments of the present invention have been described above, those skilled in the art will appreciate that the spirit and scope of the present invention are not limited to the specific description and drawings in this specification, and are not limited thereto. It goes without saying that it is possible to cover various modifications and changes all set forth in the claims.
[0051]
【The invention's effect】
As described above, according to the present invention, in a document in which a text and an image are mixed, the text is summarized, and the image is selected or reduced to generate the summary document. The user can efficiently grasp the contents of the document while keeping the amount to an appropriate amount.
[Brief description of the drawings]
FIG. 1 is a block diagram illustrating a basic configuration of a document processing apparatus according to a first embodiment of the present invention.
FIG. 2 is a block diagram illustrating a specific configuration of the document processing apparatus according to the first embodiment of the present invention.
FIG. 3 is a flowchart illustrating a flow of a processing operation of the document processing apparatus according to the first embodiment of the present invention.
FIG. 4 is a diagram illustrating a specific example of the document processing apparatus according to the first embodiment of the present invention.
FIG. 5 is a diagram illustrating a specific example of the document processing apparatus according to the first embodiment of the present invention.
FIG. 6 is a diagram showing an example of the contents of a text element holding unit in the document processing device according to the first embodiment of the present invention.
FIG. 7 is a diagram showing an example of the content of an image element holding unit in the document processing device according to the first embodiment of the present invention.
[Explanation of symbols]
101 Document holding unit 102 Component extraction unit 103 Text element holding unit 104 Image element holding unit 105 Text summarization processing unit 106 Image selection unit 107 Summarized document synthesizing unit 108 Summarized document holding unit 109 Summarized document output unit 201 CPU
202 memory 203 control memory 204 output device 205 bus

Claims (15)

Component extracting means for extracting at least a text element and an image element from the document,
Summary text generating means for generating a summary text from the text element;
Summary image generation means for generating a summary image from the image element;
A document processing apparatus comprising: a summary document generation unit configured to generate a summary document from the summary text and the summary image. 2. The document processing apparatus according to claim 1, wherein said summary image generation means selects an image element referred to from said summary text to generate a summary image. 2. The document processing apparatus according to claim 1, wherein the summary image generation unit selects an image element belonging to the same unit as the summary text to generate a summary image. Text similarity determining means for determining similarity between texts is further provided,
2. The summary image generation unit according to claim 1, wherein the summary image generation unit selects an image element to which a description similar to the summary text is provided from the image elements to generate a summary image. 3. Document processing device. The document is an electronic document;
5. The method according to claim 1, wherein the component extracting unit extracts the text element and the image element by analyzing structural information of the digitized document. Document processing device. The document is a document image;
2. The method according to claim 1, wherein the component extracting unit includes an area separating unit that extracts a text region and an image region from the document image, and a character recognizing unit that extracts a text character string from the text region. The document processing device according to any one of claims 1 to 4. The summary image generating means includes image element reducing means for generating a reduced image of the image element, and image element selecting means for selecting a part of the image element. 2. The document processing apparatus according to claim 1, further comprising a mode selection unit for selectively selecting a second mode using at least one of these two units. A component extraction step of extracting at least a text element and an image element from the document,
A summary text generating step of generating a summary text from the text element;
A summary image generating step of generating a summary image from the image element;
A document processing method, comprising: generating a summary document from the summary text and the summary image. 9. The document processing method according to claim 8, wherein the summary image generating step is to generate a summary image by selecting an image element referred to from the summary text. 9. The document processing method according to claim 8, wherein in the summary image generating step, a summary image is generated by selecting image elements belonging to the same structural unit as the summary text. Further comprising a text similarity determination step of determining similarity between texts,
9. The summary image generating step according to claim 8, wherein, in the image elements, an image element to which an explanation similar to the summary text is added is selected to generate a summary image. Document processing method. The document is an electronic document;
13. The method according to claim 8, wherein the component extracting step extracts the text element and the image element by analyzing structural information of the digitized document. Document processing method. The document is a document image;
2. The method according to claim 1, wherein the component extracting step includes an area separating step of extracting a text area and an image area from the document image, and a character recognition step of extracting a text character string from the text area. 13. The document processing method according to any one of 8 to 12. The summary image generation step includes an image element reduction step of generating a reduced image of the image element, and an image element selection step of selecting a part of the image element, and a first mode in which both steps are used. 9. The document processing method according to claim 8, further comprising a mode selection step of alternatively selecting a second mode using at least one of these two steps. A control program for a document processing apparatus, comprising a program code for causing a computer to execute each step of the document processing method according to any one of claims 8 to 14.

Images