Document processing system.

0001The present invention relates to a document processing system which can automatically lay out image data such as figures, tables, and the like at appro­priate positions in document data.

0002In a conventional document processing system such as a wordprocessor, when image data such as figures and/or tables is laid out and embedded in a document created by the system, for example, a blank space is inserted in text data constituting a document, and image data of a figure and/or table is attached to the space. With this method, however, each time document data is edited later, the layout operation must be performed.

0003A function of setting a blank region in document data, e.g., a blank-open function or a region designa­tion function, is provided to a wordprocessor. With this function, the image region is permanently laid out to a specific portion in a specific page. However, in this method, when document data is edited after the image region is laid out, a reference portion of a figure and/or table in text data is considerably separated from the fixed image region in a page, and the correspondence therebetween cannot be easily grasped. Such a case occurs when document data before the refer­ence portion of the figure and/or table is greatly increased or decreased.

0004In this manner, when a blank space is formed at a given position on a page, if document data is edited thereafter, the position or size of the space in text data must be changed.

0005In contrast to this, blank lines are inserted in text data to keep an image region in text data. In this case, if document data is edited, the blank space is moved together with the text data, and the reference portion of a figure and/or table in document data can be prevented from being separated from a layout position of the figure and/or table. In this case, however, if these blank lines extend across two pages, the image region also extends across two pages, and the positions of the blank lines must be changed.

0006When a document is created using a wordprocessor, image data of an existing figure and/or table is often merged in document data. In this case, the size of image data is checked beforehand, and a space corre­sponding to the image size must be inserted in the document data, resulting in very cumbersome processing. If the size of the image data is changed, the size of the space in the document data must be changed.

0007In the conventional system as described above, (1) if an image region is permanently set at a position in a given page of document data, the relation between a text and a figure and/or table is often changed by editing of document data, and a cumbersome operation is required to obtain the correct relation. (2) If blank lines are inserted in text data, a set of blank lines for laying out a figure and/or table often extends across two pages by editing of document data, and a cumbersome operation is required to obtain the correct relation. (3) When a space for fetching image data is set in document data, the size of the image data must be checked, and this operation is also cumbersome. (4) When the size of image data is changed, the size of a space in document data must also be changed.

0008It is an object of the present invention to provide a document processing system which does not require, to an editor, i.e., an operator, a cumbersome operation when document data or image data is edited, and which can automatically and appropriately lay out an image region in document data.

0009A document processing system of the present inven­tion comprises an input section for inputting text data and image data constituting document data, the text data including key information corresponding to the image data, and the image data being laid out in document data; a memory section for storing the text data and the image data input at the input section; a text analyzing section for specifying a position in the document data at which the image data is to be laid out, based on the position of key information in the text data in the document data; an image selecting section for specifying image data corresponding to the key information based on the key information with reference to image data stored in the memory section; an image size identifying section for identifying the size of the image region of the image data specified by the image selecting section; a layout processing section for laying out the image region of the selected image data in accordance with a predetermined layout rule; and an output section for outputting document data in which the image region is laid out by the layout processing section.

0010According to the present invention, a position at which image data in document is to be inserted is detected based on key information, such as a number or title of a figure and/or table, which corresponds to image data such as a specific figure and/or table, included in text data. Image data to be inserted is specified in accordance with the key information. When the insertion position and content of the image data are specified, an image region having a size designated by the image data is laid out in document data. In this case, an image data layout operation is performed in accordance with a predetermined layout rule, and layout of an image region extending across two pages or separa­tion of an image reference portion in text data from the corresponding image data is avoided.

0011Therefore, according to the present invention, when document data, i.e., text data and image data, are edited later, the image data can be laid out to an optimal position without requiring any load to an editor.

0012According to the present invention, if text data and image data constituting document data are separately stored, key information for specifying image data and its insertion position is extracted from the text data. The image data can be specified and the size thereof can be identified based on this information. Thus, the image data can be laid out in document data without extending across two pages of document data. Therefore, a user (operator) need not perform an operation for forming a space corresponding to a region size of image data, and need not care if image data extends across two pages. Since automatic layout operation can be perform­ed, a user's load can be greatly reduced, and effective document processing can be performed.

0013This invention can be more fully understood from the following detailed description when taken in conjunction with the accompanying drawings, in which: <ul id="ul0001" list-style="none"><li>Fig. 1 is a block diagram showing a schematic arrangement of a document processing system according to an embodiment of the present invention;</li><li>Fig. 2 is a view showing an example of text data input to the system shown in Fig. 1;</li><li>Fig. 3 is a view showing a storage content of image title dictionary in the system shown in Fig. 1;</li><li>Fig. 4 is a view showing structural analysis of an image title in the system shown in Fig. 1;</li><li>Fig. 5 is a view showing an example of image data input to the system shown in Fig. 1;</li><li>Fig. 6 is a flow chart showing image title analysis processing in the system shown in Fig. 1; and</li><li>Figs. 7A to 7H are views for explaining two differ­ent printing images of document data consisting of text data and image data in the system shown in Fig. 1.</li></ul>

0014An embodiment of the present invention will be described with reference to the accompanying drawings.

0015Fig. 1 is a block diagram showing a schematic arrangement of a document processing system according to an embodiment of the present invention.

0016Input section 1 is used for inputting text data constituting a document, image data such as figures and/or tables, and commands for an edit operation. Text data memory 2 stores the text data input through input section 1. Image data memory 3 stores the image data input at input section 1. Text data analyzing section 4 analyzes the text data stored in text data memory 2 in units of, e.g., sentences to detect predetermined key information in the text data, thereby specifying a position of the key information as a position at which a figure and/or table is to be inserted. The key infor­mation is information for referring to or indicating a figure and/or table. Note that in this embodiment, a title of a figure or table, i.e., an image title is used as key information. The image title is extracted by image title extracting section 5 with reference to image title dictionary 6. Document structure memory 7 stores the text data analyzed by text data analyzing section 4 and the key information position obtained as a result of analysis. Image data identifying section 8 retrieves image data having the same image title as that extracted from the text data from image data memory 3, and identi­fies it. Image size identifying section 9 identifies the region size of the identified image data. Open &amp; layout controller 10 performs layout-control of the image data having the identified size at a position corresponding to the key information position stored in document structure memory 7. The layout control is performed in accordance with an image layout rule stored in layout rule memory 11 while opening or developing text data in a page image. Output section 12 outputs the document data in which image data is laid out. Document processing controller 13 controls the opera­tions of these sections.

0017Text data input from input section 1 is stored in text data memory 2 as original text data and is also sent to text data analyzing section 4 under the control of document processing controller 13. Text data analyz­ing section 4 analyzes input text data in units of sentences divided by a period or a return code, and extracts a sentence serving as an image title. For example, assuming that text data shown in Fig. 2 is input, matching processing between sentences and image title dictionary 6 shown in Fig. 3, i.e., processing for determining whether sentences match with the contents of image title dictionary 6, is performed. As a result, a sentence [Fig. 2 Relation Between Text and Image Region] in text data shown in Fig. 2 is analyzed into three categories, i.e., an image classification word, an alphanumeric expression, and a noun phrase, as indicated by A in Fig. 4. In some cases, an image title may include an additional word such as "No." in addition to the above-mentioned three categories like the content of image title dictionary shown in Fig. 3. A category pattern consisting of the categories is analyzed in accordance with an image title rule prestored in image title extracting section 5, and this sentence is analyzed to be an image title, as indicated by B in Fig. 4. The value of key information position indicated by, e.g., a sentence number of the sentence, and a structural attribute indicating the image title are stored in document structure memory 7. The image title rule is as follows.

Image Title Rule

[Rule 1]

0018Condition 1. A sentence is constituted in the order of an image classification word, an alphanumeric expression, and a noun phrase.     [Result] The sentence is an image title.

[Rule 2]

0019Condition 1. A sentence is constituted in the order of an image classification word, an additional word, an alphanumeric expression, and a noun phrase.     [Result] The sentence is an image title.

[Rule 3]

0020Condition 1. A sentence is constituted in the order of an additional word, an alphanumeric expression, an image classification word, and a noun phrase.     [Result] The sentence is an image title.     ...(others)

0021Image data identifying section 8 performs analysis of image data shown in Fig. 5 as in text data analyzing section 4. More specifically, section 8 extracts and identifies an image title in the image data. This identification process is performed in the same manner as in image title identification in text data described above using the image title rule described above with reference to image title dictionary 6 having the content shown in Fig. 3. As a result, if image data including an image title of the same category pattern as that stored in document structure memory 7 is detected, the image data is identified as one to be inserted at key information position P. Fig. 6 shows this processing flow. More specifically, image data is read out from image data memory 3 (step 21). A sentence until a period or a carriage return is extracted (step 22). It is checked if the extracted sentence coincides with the content of image title dictionary 6 (step 23). Matching processing (step 24) for determining whether or not the category pattern of the sentence coincides with any of the image title rules, and matching processing (step 25) for determining whether or not the category pattern of the image title extracted from the image data coincides with the category pattern of the image title extracted from the text data, are sequentially performed. As a result of these matching processing operations, if a coincidence is found, the corresponding image data is transferred to image size identifying section 9 (steps 26 and 27).

0022Image size identifying section 9 identifies the size of the image data. The size of the image data is, for example, a portion indicated by a broken line in Fig. 5, and the size information is stored in image data memory 3 to be paired with the image data. Document processing controller 13 writes the identified image size at key information position P indicated by the sentence number in document structure memory 7.

0023When the relation between the image title in the text data and the image data, and its image size are extracted, open &amp; layout controller 10 develops the text data and image data in document data of a printing image. The text data is sequentially read out from the beginning of the sentence, and is laid out in the printing image as shown in any one of Figs. 7A to 7H. Note that Figs. 7A to 7H illustrate examples wherein data is developed in two columns per page. The struc­tural attribute of each sentence in document structure memory 7 is then referred. When a Pth sentence is read, it can be detected that the structural attribution of the sentence is an image title. In this case, the size of the image data is checked, and the image layout rule is read out from image layout rule memory 11 to check a layout position. Image layout rules stored in image layout rule memory 11 are as follows.

Image Layout Rule

[Rule 1]

0024<ul id="ul0002" list-style="none"><li>Condition 1. An X-size of an image region exceeds an X-size of a text frame. <ul id="ul0003" list-style="none"><li>Condition 1.1. Key information of image data is located in a lower half in the Y direction of the text frame.       [Result] The image region is arranged at the lowermost position of the frame.</li><li>Condition 1.2. Key information of image data is located in an upper half in the Y direction of the text frame.</li><li>Condition 1.2.1. A frame has a plurality of columns, and a blank frame is present to the right of the corresponding frame.       [Result] The image region is arranged at the uppermost position of the right blank frame.</li><li>Condition 1.2.2. A frame has one or a plurality of columns, and there is no blank frame to the right of the corresponding frame.       >[Result] The image region is arranged at the uppermost position of the starting frame in the next page.</li></ul></li><li>Condition 2. An X-size of an image region is equal to or smaller than an X-size of the text frame. <ul id="ul0004" list-style="none"><li>Condition 2.1. Y-size yn of image data is equal to or smaller than size yf of the remaining portion of the corresponding frame.       [Result] The image region is arranged below a sentence of sentence number P-1.</li><li>Condition 2.2. Y-size yn of image data is larger than size yf of the remaining portion of the correspond­ing frame.</li><li>Condition 2.2.1. A frame has a plurality of columns, and a blank frame is present to the right of the corresponding frame.       [Result] The image region is arranged at the uppermost position of the right blank frame.</li><li>Condition 2.2.2. A frame has one or a plurality of columns, and there is no blank frame to the right of the corresponding frame.       [Result] The image region is arranged at the uppermost position of the starting frame in the next page.</li></ul></li><li>Condition 3. Y-size yn of image data is larger than Y-size yf of a frame or Y-size yp of a page.       [Result] Y- and X-sizes of the image region are reduced to (yf/yn) or (yp/yn), and the image region is arranged.</li></ul>

0025In the case of Fig. 7A, condition 2.1 of rule 1 is applied, and an image region is laid out to the left one of the two columns. If Y-size yn of image data is larger than remaining size yf of the left frame size, as shown in Fig. 7B, image data cannot be inserted, and condition 2.2.1 of rule 2 is applied, and the P+1 sentence is read out and developed after the P-1 sentence but prior to the P sentence. Image data corresponding to an image title of sentence number P is developed in a right frame. More specifically, sen­tences are reordered, so that an image data region does not extend across two frames. These processing opera­tions are performed by open &amp; layout controller 10. After data is developed in a printed image, the obtained data is output to output section 12.

0026As shown in Fig. 7C, distance py from the upper end of the frame to key information position P is larger than a size half Y-size yf of the frame [py ≧ (yf/2)], condition 1.1. is applied, and an image region is arranged at the lowermost position of the frame, as shown in Fig. 7D.

0027Opposite to Fig. 7C, as shown in Fig. 7E, when distance py from the upper end of the frame to key information position P is smaller than a size half Y-size yf of the frame [py ≦ (yf/2)], condition 1.2.1. is applied, and an image region is arranged at the uppermost position of the right frame, as shown in Fig. 7F.

0028When Y-size yn of the image region exceeds Y-size yf of the frame or Y-size yp of a page, as shown in Fig. 7G, condition 3 is applied, and the image size is reduced to [(yf/yn) or (yp/yn) times] so that the image region falls within the frame, and the reduced image region is arranged in the frame, as shown in Fig. 7G.

0029As described above, text data and image data are separately created, and image data can be merged in text data at a position near the image title described in the text data so that an image data region does not extend across two pages, thus allowing effective document processing.

0030The present invention is not limited to the above embodiment. In the above embodiment, the same image title as that written in image data is written in text data, and is used as key information to determine the relation by matching processing. However, matching processing may be performed using only an image classi­fication word and an alphanumeric expression as key information. Thus, an image title itself need not be written in text data. For example, "Fig. 2" is extracted from text data "as shown in Fig. 2", and its insertion position can be identified. At this time, the insertion position of image data may be laid out to an identical page in which "Fig. 2" is developed, in accordance with a layout rule.

0031The size of image data is not limited to a size smaller than one page, but may be constituted by one document consisting of a plurality of pages or a plurality of documents. That is, image data need only be identified from document data using a file name or the like. In this case, a file name of image data may be inserted as background data of text data, i.e., part of control data group corresponding to text data.

0032Furthermore, the user can alter the image title rule and the image layout rule, and can set new rules in place of the rules described above.