JPH0612540B2 - Document creation support device - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a document creation device such as a word processor or a workstation, and more particularly, to an electronic document layout having a complicated layout structure in which characters, figures, images and the like are mixed. The present invention relates to a document creation support device for creating a document.

[Conventional technology]

When a document is created by a document creation device such as a word processor or a workstation, in one document, in addition to characters, vector display figures defined by vector information such as straight lines and curves, or read by an image reader. In some cases, it is desirable to mix bitmap images that are defined by bit patterns such as different images. When creating a document in which characters, vector display graphics, bitmap images, etc. are mixed, the data processing mode in the device for each of the characters, vector display graphics, and bitmap images is different. Therefore, in order to improve the processing efficiency, special areas called character frames, graphic frames, and image frames are set in the document, and when creating or editing the document, the attributes of the processing target area are determined. The processing according to the attribute is performed. In the present specification, the term image includes a character, a vector display figure, a bitmap image, etc. in a broad sense, and only a bitmap image in a narrow sense.

When creating a document in which characters, vector display graphics, bitmap images, etc. are mixed, the document creator sets a frame with the corresponding attribute in the document while considering the layout, and adjusts the size and position. I had to do some work.

[Problems to be Solved by the Invention]

However, it is necessary to set and adjust this frame for each of the character frame, the graphic frame, and the image frame.
The same work had to be repeated when there were multiple frames with the same attribute. For this reason, there has been a problem that the work of creating the document is complicated and time-consuming. In particular, even if a similar document already exists and wants to use the same layout, the creator needs to input all of them, which is not only labor-intensive but also mentally painful.

As a method for improving this, it is possible to create various kinds of documents with a basic layout in advance, search and copy them whenever necessary, and enter the contents.

However, with this method, it is necessary to register necessary layouts in advance, and the labor of initial input work is very large. In addition, it is necessary to register all documents of the required layout, which requires a huge storage capacity. When using these registered layouts, it is necessary to search for the target layout, but if there are many registered layouts, there is a problem that the search cannot be performed easily.

The present invention is to solve the above problems,
When there is a layout document similar to the one to be created, layouts of various frames are automatically generated on the spot to improve the efficiency of document creation.

[Means for Solving the Problems]

In order to achieve the above-mentioned object, the document creation support apparatus of the present invention sets different areas according to the attributes of the images included in the document to be created, and performs different processing for each area. A document creation support device used in a document creation device for performing, the image input unit for inputting a wrinkled document as an image, and the attribute of the image of the input document are discriminated, and based on the difference between these attributes. A layout analysis unit that extracts layout information indicating the size, position, and attribute of each area, and instructs the document to be created based on the obtained layout information to set the same area as the wrinkled document. And a layout generation instructing unit for performing the layout generation.

The layout analysis unit is a region analysis unit that extracts a character region, a graphic region, and an image region from an image of a document, a character unit analysis unit that analyzes attributes of a character unit in the character region, and a graphic in the graphic region. It can be configured by a graphic part analysis unit which analyzes a part to extract a line component in the graphic and outputs it as vector data.

[Action]

In the present invention, when setting the layout of the document, the image of the wrinkled document is input, and the layout analysis unit identifies the type of the area. For example, the circumscribed rectangle of the connected image included in the document is obtained, and the character area, the graphic area, and the image area are discriminated based on the size and arrangement of the circumscribed rectangle. Based on these analyzed regions, a region corresponding to a newly created document is set by an instruction from the layout generation instruction unit. Therefore, each area is set in the newly created document in the same layout as the wrinkled document.

〔Example〕

Hereinafter, features of the present invention will be specifically described based on embodiments with reference to the drawings.

FIG. 1 shows the configuration of a document creation device such as a word processor to which the document creation support device of the present invention is applied. The document creation support apparatus 1 includes an image input unit 2 for inputting a document as an image,
The layout analysis unit 3 extracts the layout of the document from the image of the input document, and the layout generation instruction unit 4 instructs the generation of a figure frame, a character frame, an image frame, etc. according to the analyzed layout. The layout designated by the layout generation instructing section 4 is expressed by the layout allocating section 5 in a form actually visible to the user. Further, the data input from the keyboard / mouse 6 is converted into a character or a graphic by the character / graphic input unit 7 and taken in,
In the content allocating section 8, the content is input into each of the previously laid out frames and displayed on the display section 9. Further, when the layout is manually input from the keyboard / mouse 6, the layout coordinate / attribute input unit 10 converts the data input from the keyboard / mouse 6 and sends it to the layout generation instruction unit 4.

FIG. 2 shows the configuration of the layout analysis unit 3. The image of the document input from the image input unit 2 is divided into regions of characters, vector display graphics, and bitmap images in the region analysis unit 11, and in the character region, the character unit analysis unit 12
Data (character frame / character part attribute data) such as the location of the character area, the size of the character, the space between the character strings, and the direction of the character string is extracted. In the figure area, the figure part analysis unit 13 detects vertical and horizontal lines such as a frame of a figure, a frame of a table, a ruled line, and a partition line of a step, and outputs them as vector data (figure frame / ruled line data). Further, the image part is output as frame data.

Next, the operation of the layout analysis unit 3 described above will be described.

First, a wrinkled original image is input to the image input unit 2 (see FIG. 1), and the binarized document image is analyzed by the area analysis unit 11. The area analysis method is not particularly limited, but as one method, there is a method of taking circumscribed rectangles surrounding the connected images and classifying the circumscribed rectangles according to size and arrangement.

FIG. 3 shows an example of the procedure of area analysis.

First, a circumscribed rectangle is extracted (step 101). On this occasion,
The circumscribing rectangles that overlap each other are replaced with a circumscribing rectangle that includes the whole and integrated (step 102). If the vertical or horizontal length of the circumscribed rectangle exceeds the size of the character, it can be identified as a bitmap image or a vector display graphic (step 103). For example, when dealing with characters up to 36 points, if the height or width is larger than about 13 mm,
Considered as a bitmap image or vector display figure. Further, among the large circumscribed rectangles, those having an aspect ratio close to 1 and a high ratio of black pixels in the circumscribed rectangle can be identified as an image area (step 104). In addition, when the aspect ratio is close to 1, it means 1 /
It means in the range of 3 to 3. Other large bounding rectangles can be considered graphic regions. Further, the small circumscribed rectangle is determined as a character, but it can be confirmed by further examining the periodicity of characters in the horizontal and vertical directions (step 105). The set of circumscribing rectangles having a substantially constant horizontal / vertical character cycle obtained at this time is extracted as a character group having the same attribute, that is, a character area (step 10).
6). By the above processing, the character area, the graphic area, and the image area are extracted, and the size / position information of each area is obtained.

In the above-mentioned area analysis, each area was identified by the size of the circumscribed rectangle, the aspect ratio, etc. In addition to this, Kida et al. "Method of constructing an automatic document recognition system", The Institute of Image Electronics Engineers, Vol. An algorithm using the marginal distribution method and the black concatenation method as described in No. (1986), P107-115 can be adopted.

Next, the character part analysis unit 12 will be described. Here, the internal structure of the area extracted as the character area by the area analysis unit 11 is analyzed. Here again, the method is not particularly limited. As one method, the circumscribing rectangle extracted previously and its period are used. An example of the general procedure is shown in FIG.

First, the direction of the character string is detected. Usually, the character spacing is narrower than the character string spacing. Therefore, the interval between the rectangles is extracted (step 201), and it is determined that the direction in which the average interval of the circumscribing rectangles in the horizontal and vertical directions is small is the direction in which the character strings are arranged (step 202). That is, when the vertical interval is larger than the horizontal interval, it is determined to be horizontal writing and the character string interval is set to the vertical cycle (step 203), otherwise it is determined to be vertical writing and the character string interval is set to the horizontal cycle (step 203). Step
204).

Next, the size of the character is obtained from the maximum values of the width and height of the circumscribed rectangle in the area (step 205). with this,
The attribute in each character area can be specified.

Next, in the figure part analysis section 13, the area determined as the figure by the area analysis section 11 is analyzed, and the line figure is vectorized. It is also possible to vectorize all the detected line figures, but here, because of structural importance such as frame information and ruled lines, a table frame consisting of vertical and horizontal straight lines, partition lines of columns, etc. Only long vertical and horizontal line segments that are considered to be highly useful are extracted. FIG. 5 shows an example of the analysis procedure.

As a method of extracting vertical and horizontal lines, black pixels are traced horizontally and vertically, and only objects having a predetermined length or more are left (steps 301 and 302). After that, the obtained vertical and horizontal lines are converted into start point and end point information and vector information indicated by thickness and the like (step
303). Of these vectorized lines, the one in which only one line segment exists independently is regarded as a ruled line or a dividing line of columns (step 304). Also, the part that is composed of a combination of multiple horizontal and vertical lines is considered to be the table part (step
305), if one rectangle is composed of horizontal and vertical lines, it is regarded as a frame of a figure (step 306). It should be noted that anything other than this is regarded as a general figure.

With the above, the layout analysis is completed, and the image input unit 2
The image of the original read in is divided into a character area, an image area, and a graphic area. The graphic area also includes tables and lines. Further, in the character area, attributes such as internal character size, character string interval, and character string direction are added. Based on these pieces of information, the layout generation instructing section 4 shown in FIG. 1 instructs the layout allocating section 5 to perform layout and allocates each area.

For example, a wrinkled document including a character portion a, a bitmap image portion b, a graphic portion c, a partition line d, a table e, etc. as shown in FIG. 6 (a) is read by the image input unit 2. In this case, if you perform layout analysis on this image,
The result shown in (b) is obtained. In the figure (b), A
Is a character area, B is an image area, C is a graphic area, D is a partition line, and E is a table structure. Since each area based on the layout analysis is displayed on the screen of the display unit 9 on which the document is displayed, the document creator designates the area where the character or the figure should be input by the keyboard / mouse 6, and the keyboard /
A document having a desired layout can be generated only by inputting desired contents in each area using the mouse 6.

When the bitmap image is directly embedded in the document, the image data input from the image input unit 2 may be sent to the content allocating unit 6 and combined.

Further, the document creator may make a correction to edit each area generated by the layout analysis and then input the content.

In the above-described embodiment, the document creation is described as an example, but the invention can be applied not only to the document but also to the creation of a predetermined format such as a slip.

Further, in the present embodiment, the table portion only reproduces the vertical and horizontal lines based on the vector data obtained by the analysis, but if the generation and management of the table structure is possible on the document creation device side, Instead of directly outputting vector data,
It is also possible to output as information indicating the attributes of the table, which is the number of rows and columns.

Further, in the above-described embodiment, the information extracted in the character portion is only the character size, character string interval, vertical writing and horizontal writing, but the typeface difference and the language difference, for example, the difference between Japanese and English. May be identified and these attributes may be added.

Further, by recognizing characters in the character part and recognizing figures in the graphic part, not only the layout information but also the content itself can be reused.

〔The invention's effect〕

As described above, in the present invention, since the layout is automatically generated based on the read image of the original, the document creator sets each area when creating a document having a similar layout. The work of making corrections is unnecessary or simple, and the efficiency of document creation is improved. Further, since it is not necessary to create or register the standard layout document in advance, the procedure for the preparation stage is unnecessary, and a large-capacity storage device is not required.

[Brief description of drawings]

FIG. 1 is a block diagram showing the configuration of a document creation device to which the document creation support device of the present invention is applied, FIG. 2 is a block diagram showing the configuration of a layout analysis unit, and FIG. 3 is an example of a region analysis procedure. FIG. 4 is a flowchart showing an example of a procedure of character part analysis, FIG. 5 is a flowchart showing an example of a procedure of graphic part analysis, and FIGS. 6 (a) and 6 (b) are a manuscript image and layout analysis. It is explanatory drawing which shows a result example typically. 1: Document creation support device 2: Image input unit 3: Layout analysis unit 4: Layout generation instruction unit 5: Layout allocation unit 6: Keyboard / mouse 7: Character / figure input unit 8: Content allocation unit 9: Display unit 10: Layout coordinate / attribute input section 11: Area analysis section, 12: Character section analysis section 13: Graphic section analysis section

Claims (2)

[Claims] 1. A document creating apparatus used in a document creating apparatus, wherein different areas are set according to attributes of an image included in a document to be created and different processing is performed on each area when creating a document. An image input unit for inputting a wrinkled document as an image, which is a support device, and an attribute of an image of the input document is discriminated, and the size, position, and attribute of each area are determined based on the difference of the attributes. A layout analysis unit that extracts layout information that indicates the layout information, and a layout generation instruction unit that instructs the document to be created based on the obtained layout information to set the same area as the wrinkled document. A document creation support device characterized by: 2. The layout analysis unit includes a region analysis unit that extracts a document region, a graphic region, and an image region from a document image; a character unit analysis unit that analyzes attributes of a character unit in the character region; 2. The document creation support apparatus according to claim 1, further comprising a graphic part analysis unit that analyzes a graphic part in the graphic area to extract a line component in the graphic and outputs it as vector data.