JPH0981583A - Device and method for document filing - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide the device and method for document filing which facilitate registration and retrieval of a document picture in a filing device. SOLUTION: With respect to the document filing device which takes in pictures of various documents to file them, the picture of a document is inputted by a picture input part 11 and is stored in a memory 12, and a picture processing part 13 segments a desired partial picture from the entire picture of the document stored in this memory 12, and the entire picture and the partial picture are preliminarily stored as document data in a preservation part 14 correspondingly to each other, and this partial picture is shown in a display part 15 as the index at the time or retrieval of the document.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a document filing apparatus and method for filing various documents, and a document filing apparatus and a method for facilitating registration and retrieval of a document image in a file apparatus only by adding a simple structure, Regarding the method.

[0002]

2. Description of the Related Art In a conventional document filing apparatus, as a method of saving a document image, a method of capturing an image of one document and giving a name to this file and saving it, and characters and figures described on the document A method of converting the image information into code information instead of image information and saving it as a code file is often used. Further, in recent years, a method of partially storing an image of a document instead of storing the entire image of a document has been proposed. For example, JP-A-4-321188
The publication discloses a method of storing only the image entered in the form by using the format data of the blank form.

Further, in the conventional document filing apparatus, as a method of searching the stored document image,
A method of designating one of the keywords set and presented in advance and a method of allowing an operator to input an arbitrary keyword from a keyboard or the like are often employed.

However, in the method of storing and managing the image of one document, it is necessary for the operator to manually give a unique name to each image as keyword information for the search. Was extremely complicated.

On the other hand, the characters written on the document,
In the method of converting a figure from image information to code information, the code data after conversion can be used as search information, but many of these conversion operations need to be performed manually by an operator, which is extremely labor intensive. It's a lot.

Further, the conventional retrieval method has a problem that a huge amount of time is spent when a filed document is displayed and retrieved, if the amount of the form is large.

[0007]

As described above, in the conventional document filing apparatus, one image of a document is named and saved, or the document is converted into code information and saved. There are things, but in any case, before searching for a document, it is necessary for the operator to manually give the information to be used to the document in advance,
There is a problem that the work is remarkably complicated and requires a lot of labor.

Further, in the conventional search system, when displaying and searching a filed document,
There was a problem that a huge amount of time was spent when the amount of forms was large.

The present invention has been made in view of the above circumstances, and it is an object of the present invention to provide a document filing apparatus and method that facilitate registration of a document image in a file device and facilitate document retrieval. To aim.

[0010]

SUMMARY OF THE INVENTION The present invention is a document filing apparatus for fetching and filing images of various documents, and image input means for inputting the image of the document, and the document input by the image input means. It comprises a partial image cutout means for cutting out a partial image from the entire image, and a file means for storing the entire image and the partial image in association with each other as document data, and when the document is searched, the partial image is used as an index. It is characterized by being presented to the user.

According to the present invention, when a document is registered in a file as character data and image data, a desired document to be registered is input as an image, and a partial image serving as a document index is cut out from the input image data. Save it in association with the document. Then, this partial image is presented to the user and used for document retrieval.

That is, according to the present invention, when a document is saved in a file, the partial image index can be automatically added and saved, which facilitates registration of the document in the file and saves the partial image. The documents in the used files can be easily searched.

The partial image cutout means may be a means for cutting out the partial image from the entire image based on a user's instruction or preset cutout information.

Further, it is preferable that the file means includes means for associating a plurality of partial images with the whole image, and when the document is searched, the plurality of partial images are sequentially presented to the user as an index.

Further, the partial image cutout means has a plurality of kinds of cutout patterns instructed by a user or the cutout information, and the file means is the cutout partial image and the cutout obtained by cutting out the partial image. It is preferable to include means for storing the pattern types in association with each other.

The file means further comprises a recognizing means for performing character recognition or graphic recognition on the partial image associated with the whole image and converting the partial image into code information. It is preferable to include means for storing in association with each other.

Further, the partial image cutout means includes a line segment extraction means for extracting a line segment from the whole image and a character string extraction means for extracting a character string area from the whole image, and the extracted line segment. It is preferable to cut out the image of the character string region existing above.

Further, it is preferable that the partial image cutout means includes a modified character extraction means for extracting a modified character area from the entire image, and cuts out an image in the extracted modified character area.

As the modified character extraction means, a character frame extraction means, a character string extraction means, a line segment extraction means, and a line width, a black pixel density, a contour line segment or a connected area of black pixels in the area are measured. It is characterized in that at least one of the measuring means is included and at least one of a shaded character area, an underlined area, an italic character area, an outlined character area, an enlarged character area, and a bold character area is extracted.

The partial image cutting means may include means for extracting a character string area from the entire image, and may be configured to cut out an image in the character string area existing at a position designated in advance.

Further, a recognition means for performing character recognition on the image to convert it into code information is further provided, and the partial image cutout means includes a character extraction means for extracting a character string region from the whole image. It is preferable that the character code in the extracted character string area is acquired by the character recognition means, and an image in the character string area is cut out when the acquired character code includes a preregistered character code.

Further, the image processing apparatus further comprises a recognizing means for performing character recognition on the image to convert it into code information, and the partial image cutting means includes a character extracting means for extracting a character string area from the whole image. A character code in the extracted character string area is acquired by the character recognition means, and when the acquired character code includes a preregistered character code, it is designated by position information indicating an area to be extracted as a partial image. It is more preferable to cut out the partial image existing at the specified position.

Further, the partial image cutout means extracts means for drawing, table and picture areas from the whole image,
A character string extracting unit for extracting a character string area from the entire image, and cutting out an image in the character string area located above or below the figure, table and picture areas.

Further, the partial image cutout means includes means for extracting a figure, table and picture area from the whole image, and cuts out an image of the figure, table and picture area.

Further, the file means is characterized in that the partial image cut out by the partial image cutout means is stored as either the partial image itself or position information on the entire image of the partial image.

[0026]

BEST MODE FOR CARRYING OUT THE INVENTION An embodiment of the present invention will be described below with reference to the drawings.

FIG. 1 is a block diagram showing the basic processing functions of the document filing apparatus according to the embodiment.

As shown in FIG. 1, the document filing apparatus according to the embodiment has a control unit 10, an image input unit 11, a memory 12, an image processing unit 13, a storage unit 14, a display unit 15, an input unit 16 and characters. The recognition unit 17 is provided.

The control unit 10 controls the entire document filing apparatus. The image input unit 11 also inputs images of various documents such as drawings and documents. The image input unit 11 may be one that optically inputs the document information described on the document using an image scanner such as a line sensor, or a binary image is converted from the image file. It may be one that reads the document information that is stored.

The image of the document input from the image input unit 11 is stored in the memory 12. Then, the image processing unit 13 cuts out a partial image from the image data stored in the memory 12. Further, the storage unit 14 stores the image of the input document, the cut-out image, the code data of the recognition result, and the like. Then, the display unit 15 displays the document stored in the storage unit 14 and the cut out partial image.

The input unit 16 is a keyboard, a mouse or the like used when selecting one of the documents displayed on the display unit 15. Also,
The character recognition unit 17 acquires the character code of the partial image cut out by the image processing unit 13. Therefore, the character recognition unit 17 stores the partial image as the image information in the storage unit 14.
It will not be used when holding.

Next, the operation of the document filing apparatus of the same embodiment will be described with reference to FIG.

First, the image input section 11 scans the image of the document to be filed and stores the obtained image signal in the memory 12 (step A1 in FIG. 2). Next, the image processing unit 13 cuts out a desired partial image from the image signal of the document stored in the memory 12 based on the cutout information (step A2 in FIG. 2). The cutout information may be instructed by the user via the input unit 16 or may be obtained by an automatic process described later. Then, the image processing unit 13 associates the cut-out partial image with the document and files in the storage unit 14 (step A3 in FIG. 2).

As an example of this correspondence, if the input image name is [input 001], the partial image is [input 0
01 partial image 001] and the like, it becomes possible to perform mutual searches such as a partial image from a file name and a partial image from a file name.
Any method may be used as the image filing method.

FIG. 3 shows an example in which partial images and documents are associated with each other. As for the partial image to be stored here, a method of holding a part of the input image as it is, a method of holding only the positional information on the input image, and the like can be considered.

On the other hand, when searching a document filed in the storage unit 14, as shown in FIG.
Each partial image is displayed as a list on the display unit 15 in association with an icon indicating one document. Then, the user inputs one of the displayed partial images into the input unit 16
The document is stored in the storage unit 14 by instructing via the
Load from. For example, when the partial image [partial image 001] displayed on the display unit 15 is instructed by the user via the input unit 16, the correspondence of the [input 001 partial image 001] stored in the storage unit 14 is obtained. Based on the relationship, the document [input 001] is loaded from the storage unit 14.
In addition, with respect to this displayed partial image, the storage unit 1
When there are a plurality of items held in 4, the partial images to be displayed are switched so that they will be displayed after a certain period of time, or the “next” partial image is instructed to be displayed via the input unit 16. The partial image is displayed so as to be switched. In addition to the method shown in FIG. 4, the method of displaying the partial images of the document stored in the storage unit 14 is not limited to that shown in FIG.
It may be displayed side by side with the creation date and keywords.

FIG. 6 shows an operation procedure for cutting out the character string image located on the line of the embodiment.

This clipping is based on the possibility that the character string image located on the line can be a partial image representing the entire document.

The image processing section 13 first extracts a line segment in the image (step B1 in FIG. 6). This line segment extraction is performed, for example, by the procedure shown in FIG.

That is, first, the input data is raster-scanned to detect the tracking start point thereof, and then the contour is extracted by tracing the closed curve formed by the figure boundary clockwise or counterclockwise from the tracking start point ( Step C1 of FIG. 7). The contour information extracted by such closed curve tracing is stored as a chain code that is a sequence of coordinate points or a sequence of direction codes. It is needless to say that other contour extraction methods conventionally proposed variously may be used for this contour extraction processing.

Next, based on this contour line data, characteristic points such as intersections and corner points of the image are detected. For example, the characteristic points are a point sequence having a convex / concave shape from the contour point sequence. To extract. This feature point extraction is performed by, for example, comparing the curvature k of the contour point sequence obtained by the curvature calculation process (step C2 in FIG. 7) with predetermined threshold values K0 and K1, and (k ≦ K
The points 1) are obtained as a convex point sequence (step C3 in FIG. 7).

Next, the start point / end point are calculated by using these convex / concave contour point sequences as line segments. Thereby, the straight line segment portion in the input image can be detected, and the line segment longer than the threshold value T is subjected to the following processing (step C4 in FIG. 7).

The line segment extraction method may be a method of measuring a series of black pixels in the horizontal direction of an image, detecting a portion longer than a threshold T to detect a horizontal line segment, and various methods have been proposed in the past. You may use other methods, such as the method.

Next, the image processing unit 13 extracts the character string in the image (step B2 in FIG. 6). This character string extraction is performed by the following method, for example.

First, the image is labeled for each connected area to obtain a circumscribing figure. Alternatively, all contours of the black area are obtained, and a circumscribed rectangle of each contour is obtained. Next, if adjacent circumscribing figures are within a certain threshold distance, they are merged. Then, when this merging is performed, the circumscribing figure is calculated again, and if the variation in the direction of the underline is within a certain threshold value, the same character string is determined. As the character string extraction method, other methods such as variously proposed methods may be used.

Next, the image processing unit 13 associates the line segment thus extracted with the character string (step B3 in FIG. 6). This association is performed, for example, by the procedure shown in FIG.

That is, first, a line segment longer than the threshold Th is detected (step D1 in FIG. 8), and the upper or lower region of the detected line segment is searched to select the selected line segment and the baseline of the character string frame. When the distance is less than or equal to the threshold value d (step D2 in FIG. 8).

Separately from this, extraction of the character area located in the vicinity of this line segment is performed by searching the upper portion of the line segment obtained by the line segment detection shown in FIG. 7, as shown in FIG. It may be configured such that the size of the connected region of black pixels is equal to or smaller than a threshold value and these circumscribed rectangles are set as partial images.

Next, the image processing unit 13 detects the character string located on the line (step B4 in FIG. 6).

According to such a detection procedure, when the document image shown in FIG. 10 is taken as an example, the character string image as shown in FIG. 11 is cut out.

FIG. 12 shows an operation procedure for cutting out the modified character image of the embodiment.

This cutout is based on the possibility that a character string that has been modified, such as by knitting, can be a partial image representing the document.

The image processing unit 13 detects modified character areas such as a shaded character area, an italic character area, an enlarged character area and a thick line character area (step E1 in FIG. 12), and cuts out an image of the detected area ( Step E2 in FIG. 12).

An example of a method of detecting each modified character area will be described below.

Extraction of the shaded character area is performed by, for example, FIG.
Follow the procedure shown in.

First, the image processing section 13 obtains a connected area of black pixels (step F1 in FIG. 13). Then, a mark is attached to a region where the length of each side of the circumscribed rectangle of the connected region is equal to or less than the threshold value h (step F2 in FIG. 13). Next, the image processing unit 13 obtains an area in which the length of each side of the circumscribed rectangle of the connected area is less than or equal to the threshold value c, and if the distance between another such connected area and its center is less than or equal to the threshold value D. Both areas are integrated as one area (step F in FIG. 13).
3). This process is performed for all the lengths of the sides of the circumscribed rectangle of the connected area that are equal to or less than the threshold value c.

Next, the number of marked areas inside the integrated area is measured, and the value is divided by the area of the circumscribed rectangle of the integrated area to calculate a value (step F4 in FIG. 13). . Then, when this value is equal to or greater than the threshold value p, the integrated area is output as a shaded character area (step F5 in FIG. 13). Of course, any other method may be used to extract the shaded character area. For example,
The input image is first divided into square areas of size m, and the number of connected areas in which the length of each side of the circumscribed rectangle inside the square area is equal to or less than the threshold value h is measured. Then, the square area of which the number is equal to or larger than the threshold value q is marked as an area in which a shaded character area exists. Next, adjacent ones of the marked square areas are obtained, and the squares adjacent to each other are grouped as one area. The area thus created is output as an area in which a shaded character area exists.

The extraction of the italic character area is performed as follows, for example.

First, the image processing unit 13 extracts a black area forming a character. This extraction is performed, for example, by obtaining a circumscribing figure of black pixels to be connected that is equal to or smaller than a threshold. Any other method may be used to extract this character component. Next, the image processing unit 13 extracts a character string. This character string extraction is performed by integrating the character components found previously and located in the vicinity. Then, the pattern of 2 × 2 pixels in the character string region is observed, and the number of patterns shown in FIGS. 14A to 14H is obtained. Then, a value obtained by dividing this number by the number of black pixels in the character string area is calculated, and when this value is equal to or larger than the threshold value, this portion is determined to be an italic character area.

Of course, any other method may be used to extract the italic character area. For example, the input image is first divided into square areas of size m, and 2
By observing the pattern of the pixel of × 2, FIG.
The number of patterns shown in (h) is measured. Then, the ratio of this number to the black pixels inside the square area is calculated,
When this is greater than or equal to the threshold q, the square area is marked as an area in which italic characters are present. Next, in each of the marked square areas, the adjacent ones are obtained, and the squares adjacent to each other are grouped into one area. The area thus created is output as an area in which an italic character area exists.

In the extraction of the enlarged character area, the area in which the size of the circumscribed rectangle of the area obtained by the above-mentioned character string extraction is not less than the threshold value is output as the enlarged character area.

The extraction of the thick line character area is performed as follows, for example.

First, the image processing section 13 obtains a connected area of black pixels. A region in which the length of each side of the circumscribed rectangle of the connected region is less than or equal to the threshold value c is obtained. If the distance between the other connected region and its center is less than or equal to the threshold value d, both regions are regarded as one region. Integrate. This process is performed for all the lengths of the sides of the circumscribed rectangle of the connected area that are equal to or less than the threshold value c. Next, the black area inside the integrated area is converted into white pixels adjacent to the white pixel. This conversion process is repeated n times. Next, the smallest rectangle that encloses the black area of this converted image is obtained. When the difference between the length of each side of the rectangle and the length of each side of the circumscribing rectangle surrounding the first character string region is equal to or less than the threshold value, this region is determined to be a thick line character region.

As a result, modified character areas such as shaded character areas, italic character areas, enlarged character areas and thick line character areas are detected and cut out as partial images.

FIG. 15 shows an operation procedure for cutting out the designated character area image portion of the embodiment.

First, the user designates a character area to be cut out via the input unit 16 (step G1 in FIG. 15).
This is done, for example, by preparing a menu button as shown in FIG. 16A and designating the region of the character to be cut out, such as upper left, upper center, upper right, lower left, lower center and lower right. Alternatively, as shown in FIG. 16 (b), absolute coordinates such as (starting point, width, height) may be entered and designated. Of course, the coordinate values may be directly input. Further, any method such as a keyboard or voice may be used as the method of instructing. That is, it suffices if the character area to be cut out can be designated.

Then, the image processing section 13 cuts out the partial image in accordance with this instruction (step G2 in FIG. 15).

FIG. 17 shows an operation procedure for cutting out an image of a character string area including a character code already registered in the embodiment.

Here, it is assumed that there are a registration mode for registering a character code and a partial image cutout mode for cutting out a partial image.

In the registration mode, character code information for cutting out a partial image is registered. Then, the user inputs the character to be registered (step H1 in FIG. 17). This input may be performed from the keyboard or may be designated with a mouse or the like from the registered character information. Specifically, for example, the character code as shown in FIG. 18 is registered. It should be noted that this registration method and the registration method may use other means.

Next, the partial image cutout mode will be described.

First, the image processing section 13 extracts a character string (step H2 in FIG. 17). This character string detection is performed by the following method, for example.

First, the image is labeled for each connected region, and the circumscribing figure is obtained. Next, if adjacent circumscribing figures are within a certain threshold distance, they are merged. When this merging is performed, the circumscribing figure is calculated again, and if the variation in the direction of the underline is within a certain threshold value, the same character string is determined. As the character string extraction method, other methods such as variously proposed methods may be used.

Next, the image processing section 13 recognizes the character in the character string (step H3 in FIG. 17). Any existing method may be used for this character recognition means. Then, it is determined whether the registered character registered in advance in the registration mode is not in the recognized character string, and if the recognized character exists in the registered character, the partial image is displayed according to the previous registration information. Is cut out (step H4 in FIG. 17). For example, when a character string as shown in FIG. 18 is registered in the registration mode and a document as shown in FIG.
The character string shown in is cut out.

FIG. 21 shows an operation procedure for cutting out an image at a designated position in a character string area including a character code which is already registered in the embodiment.

Here, it is assumed that there are a registration mode for registering a character code and a partial image cutout mode for cutting out a partial image, as described above.

In the registration mode, character code information for cutting out a partial image is registered. Then, the user inputs the character to be registered and the position of the character to be cut out (step I1 in FIG. 21). This input may be performed from the keyboard or may be designated with a mouse or the like from the registered character information. Specifically, for example, the character code and the cutout position as shown in FIG. 22 are registered. It should be noted that this registration method and the registration method may use other means.

Next, the partial image cutout mode will be described.

First, the image processing section 13 extracts a character string (step I2 in FIG. 21). This character string detection is performed by the following method, for example.

First, the image is labeled for each connected region, and the circumscribing figure is obtained. Next, if adjacent circumscribing figures are within a certain threshold distance, they are merged. When this merging is performed, the circumscribing figure is calculated again, and if the variation in the direction of the underline is within a certain threshold value, the same character string is determined. As the character string extraction method, other methods such as variously proposed methods may be used.

Next, the image processing section 13 recognizes the character in the character string (step I3 in FIG. 21). As the means for character recognition, any existing means may be used. Then, it is determined whether the registered character registered in advance in the registration mode is not in the recognized character string, and if the recognized character exists in the registered character, the partial image is displayed according to the previous registration information. Is cut out (step I4 in FIG. 21). For example, when a character string as shown in FIG. 22 is registered, when a document as shown in FIG. 23 is input and there is a registered character [name], the character as shown in FIG. Rows will be cut out.

FIG. 25 is an example of cutting out an image in a character string located above or below the drawing, table or picture area of the same embodiment.

The image processing section 13 first performs area separation (step J1 in FIG. 25). In this area separation, elements in the image are separated into character strings and figures, tables and pictures areas. This separation includes character string extraction and drawing area extraction (see FIG. 25).
Step J2 to step J3).

The character string is extracted by the following method, for example.

First, the image is labeled for each connected region, and the circumscribed figure is obtained. Next, if adjacent circumscribing figures are within a certain threshold distance, they are merged. When this merging is performed, the circumscribing figure is calculated again, and when the variation in the direction of the underline is within a certain threshold value, the same character string is determined (step J2 in FIG. 25). In addition, the character string extraction method is
Other methods such as various conventionally proposed methods may be used. Then, when the size of the final rectangle obtained by merging is equal to or larger than a certain threshold value, it is set as a candidate for a figure, a table or a picture (step J3 in FIG. 25).

Next, the modified character string is detected by searching for a character string around the figure, table or picture (FIG. 2).
5 step J4). This is, for example,
It searches for a character string around (above or below) the table and the picture area. Specifically, if the distance between the character string closest to the upper side of the figure, table or picture area of interest and the upper side is less than or equal to a threshold value, this character string corresponds to the figure, table or picture area of interest. It is considered to be.
Similar processing is performed on the lower side. Then, the image of the character area obtained from this result is cut out (step J5 in FIG. 25). Other methods may be used for associating the character strings corresponding to the areas of the figure, table, or picture. It suffices if a character string region that modifies these drawing regions can be detected. FIG. 26 is an example of an original image, FIG. 27 is an example in which a drawing area is extracted, and FIG. 28 is an example in which a character string that modifies it is cut out.

FIG. 29 shows an example of cutting out an image of a drawing, table or picture area.

The image processing section 13 first performs area separation (step K1 in FIG. 29). In this area separation, elements in the image are separated into character strings and figures, tables and pictures areas. This separation is performed by extracting character strings and drawing areas (see FIG. 29).
Step K2 to step K3).

The character string is extracted by the following method, for example.

First, the image is labeled for each connected region, and the circumscribed figure is obtained. Next, if adjacent circumscribing figures are within a certain threshold distance, they are merged. When this merging is performed, the circumscribing figure is calculated again, and if the variation in the direction of the underline is within a certain threshold value, the same character string is determined (step K2 in FIG. 29). In addition, the character string extraction method is
Other methods such as variously proposed methods may be used. Then, if the size of the final rectangle obtained by the merging is equal to or larger than a certain threshold value, the candidate is selected as a figure, table or picture (step K3 in FIG. 29).

Then, the image area of the figure, table or picture obtained as a result is cut out (step K4 in FIG. 29). FIG.
0 is an example of the original image, and FIG. 31 is an example of cutting out the drawing area.

As a result, when saving a document in a file, a desired partial image or a highly probable partial image representing the document can be added as an index and saved, and this partial image index can be stored. If it is used, the search becomes easy.

[0093]

As described in detail above, according to the present invention,
When saving the image of the document to the file device, the partial image index can be automatically added and saved, which makes it easy to register the document in the file device and uses the partial image to save in the file device. The documents can be easily searched for, and the number of processing steps can be reduced without bothering the user.

[Brief description of drawings]

FIG. 1 is a block diagram showing basic processing functions of a document filing device according to an embodiment of the present invention.

FIG. 2 is an exemplary flowchart for explaining the operation of the document filing apparatus of the same embodiment.

FIG. 3 is a diagram showing an example in which a partial image and a document of the same embodiment are associated with each other.

FIG. 4 is a view showing a partial image display example of the same embodiment.

FIG. 5 is a view showing a partial image display example of the same embodiment.

FIG. 6 is a flowchart showing an operation procedure for cutting out a character string image located on a line in the embodiment.

FIG. 7 is a flowchart showing a flow of line segment extraction processing of the same embodiment.

FIG. 8 is a flowchart showing a flow of a process of associating a line character string of the embodiment.

FIG. 9 is a conceptual diagram showing an example of extracting a character region located in the vicinity of a line segment according to the same embodiment.

FIG. 10 is a diagram showing an example of an original image for extracting a line character string according to the first embodiment.

FIG. 11 is a view showing an example of a partial image in which a linear character string of the embodiment is extracted.

FIG. 12 is a flowchart showing an operation procedure when cutting out a modified character image of the same embodiment.

FIG. 13 is a flowchart for explaining a shaded character region extraction process of the same embodiment.

FIG. 14 is a conceptual diagram showing a pixel pattern of the same embodiment.

FIG. 15 is a flowchart showing an operation procedure when cutting out a designated character area image portion of the embodiment.

FIG. 16 is a view showing an example of a menu for designating an area of a designated character area image according to the same embodiment.

FIG. 17 is an exemplary flowchart showing an operation procedure for cutting out an image in a character string area when a character code already registered is included in the character string area according to the embodiment.

FIG. 18 is a diagram showing an example of registration of character codes of the same embodiment.

FIG. 19 is a diagram showing an example of an original image of the same embodiment.

FIG. 20 is a view showing a partial image cutout result of the embodiment.

FIG. 21 is an exemplary flowchart showing an operation procedure for cutting out an image inside a character string area at a specified position when a character code already registered is included in the character string area of the embodiment.

FIG. 22 is a diagram showing an example of registration of a character code and a cutout position according to the same embodiment.

FIG. 23 is a diagram showing an example of an original image of the same embodiment.

FIG. 24 is a view showing a partial image cutout result of the embodiment.

FIG. 25 is a flowchart showing an example of cutting out an image in a character string located above or below the drawing, table or picture area of the embodiment.

FIG. 26 is a diagram showing an example of an original image of the same embodiment.

FIG. 27 is a diagram showing an example of a result of extracting a diagram, table, or picture region of the embodiment.

FIG. 28 is a diagram showing an example of a result obtained by cutting out an image according to the same embodiment.

FIG. 29 is a flowchart showing an example of cutting out an image of a drawing, table or picture area.

FIG. 30 is a diagram showing an example of an original image of the same embodiment.

FIG. 31 is a diagram showing an example of a result of extracting a diagram, table, or picture region of the embodiment.

[Explanation of symbols]

10 ... Control unit, 11 ... Image input unit, 12 ... Memory, 13
... image processing section, 14 ... storage section, 15 ... display section, 16 ... input section, 17 ... character recognition section.

Claims (2)

[Claims] 1. A document filing apparatus for capturing and filing images of various documents, and an image input unit for inputting the images of the documents, and a portion for cutting out partial images from the entire image of the document input by the image input unit. An image cutting means and a file means for storing the whole image and the partial image in association with each other as document data, and presenting the partial image as an index to the user when searching the document. Document filing device. 2. A document filing method for capturing and filing images of various documents, wherein an entire image of the document is input, a partial image is cut out from the input entire image, and the entire image and the partial image are extracted. Is stored as document data in association with each other, and when the document is searched, the partial image is presented to the user as an index.