JPH0223468A - Filing device - Google Patents

Abstract

PURPOSE:To shorten the time necessary for document registration by automatically extracting retrieving information from image information obtained at the time of document reading, and registering the retrieving information and the image information of the document to an optical disk. CONSTITUTION:An image information read part 1 reads a document 8, and supplies image data to a key title recognizing part 2 and an image data extracting part 6. A key title recognizing part 2 checks whether or not the key title recognized beforehand exists on the document 8, recognizes a part 11 of a registered character, supplies it to a retrieving information extracting part 3, and segments the image data on the part 12 corresponding to position information. A code converting part 4 divides them in units of a font, converts them into corresponding character codes, and supplies them to a registering part 5. The image data extracting part 6 extracts the data to be a registering object out of the image data, an image data compressing part 7 supplies them to the registering part 5 in a prescribed compressed form, and registers them to an optical disk 9 correlating with the code from the code converting part 4.

Description

DETAILED DESCRIPTION OF THE INVENTION [Object of the Invention] (Industrial Application Field) The present invention relates to a filing device that automatically creates a search file from image information obtained by reading a document.

(Prior art) As one of the filing devices using optical disks,
Conventionally, the one shown in FIG. 5 has been known.

The filing device shown in this figure includes an image information reading section 102 that reads the contents of a document 101, and a text 110 read by the image information reading section 102.
an image data compression unit 10 that compresses image information of 1;
3, a search title manual input section 104 for inputting the title, etc. of the document 101; the title, etc. input from the search title manual input section 104, and the document image data compressed by the image data compression section 103; and a registration unit 107 for storing the information on the optical disc 108.

Then, the sentence 1101 to be registered is read by the image information reading unit 102, and the title etc. for this document 101 is read by the search title manual input unit 1.
04, the registration unit 107 associates these and registers them on the optical disc 108.

(Problem to be Solved by the Invention) However, in such a conventional filing device, each time a document 101 is registered, the search title manual input section 1
From 04 onwards, the titles of these sentences 1101 are manually created, so when the number of documents 101 is huge,
There was a problem that it took too much time to register this.

In view of the above-mentioned circumstances, the present invention makes it possible to automatically input the necessary search title, etc., simply by reading the document when the search title is standardized, thereby reducing the time required for document registration. The purpose of the present invention is to provide a filing device that can significantly reduce costs.

[Structure of the Invention] (Means for Solving the Problem) In order to achieve the above object, a filing device according to the present invention includes an image information reading section that reads an image of a document, and an image obtained by the image information reading section. A search title recognition unit that recognizes a search title part from information; an information import unit that imports information at a position corresponding to the search title recognized by the search title recognition unit; and an information import unit that imports information imported by the information import unit into a file. It is characterized by having a filing section for storing information, and the following.

(Function) In the above configuration, if a search title is set in advance, when sentence I is read by the image information reading section, the search title recognition section recognizes the search title portion of sentence 1, and the information is captured. The information located at the position corresponding to the search title part is taken in by the section, and the information is filed by the filing section.

(Embodiment) FIG. 1 is a block diagram showing an embodiment of a filing device according to the present invention.

The filing device shown in this figure includes an image information reading section 1, a key name recognition section 2, a search information extraction section 3, a code conversion section 4, a registration section 5, and an image data extraction section 6.
and an image data compression unit 7, and when a document 8 is inserted, the image information reading unit 1 reads the document 8 and extracts information to be searched (search target information) from the image information obtained by this reading operation. information)
is automatically extracted, and this search information and the image information of the document 8 are registered on the optical disc 9.

The image information reading section 1 is equipped with an imaging device such as COD, and when the document 8 is inserted into this filing device, it reads it and also recognizes the key name of the image data obtained by this reading operation. Part 2
and the image data extraction unit 6.

The key name recognition unit 2 includes a key name storage unit storing font data for a plurality of key names, and a matching detection unit. It is checked whether or not there is a preset key name on the document 8 by comparing the image data with a plurality of key name fonts stored in the key name storage section.

In this case, “Public Patent Publication”° is stored in the key name storage section.
With the key names “Publication”, “Patent Application”, “Name of Invention”, “Inventor”, and “Applicant/Representative” being registered, the characters shown in Figure 2 are written on the document 8. If so, the “Publication Patent Publication” “Publication”” on this document 8.
Patent”゛Name of the applied invention°“Inventor”°゛Applicant”゛°
A portion 11 in which the characters "Agent" are written is recognized.

Then, the key name recognition section 2 supplies this recognition result to the search information extraction section 3.

The search information extraction unit 3 includes a character cutting unit that cuts out characters located at positions corresponding to the portion 11 recognized by the key name recognition unit 2, and the key name recognition 8! I2
When the position information of the key name is supplied from , the character at the position corresponding to this position information, for example, the second
When characters as shown in the figure are written, the image data in the portion 12 corresponding to each key name is cut out and supplied to the code converter 4.

The code conversion unit 4 is equipped with a font/code conversion unit, and when the image data of each portion 12 is supplied from the search information extraction unit 3, it divides it into font units and converts these division results into correspondences. This conversion result is supplied to the registration unit 5.

In this case, the key names "Public Patent Publication""Publication""Patent Application° "°°° ° Invention""Inventor""Applicant" are registered in the key name storage section. If the characters shown in FIG. 2 are written on the document 8, then the text "Public Patent Publication""May 7, 1988" and "Voice Recognition Device" 1986-2473 on the document 8 is written. ”1986 (1986) 1
On the 13th of May, the characters ``Kazuo Yamada'', ``Tenko Kikai Co., Ltd.'', and ``Patent Attorney Shimoyamahata'' were cut out, converted into codes, and supplied to the registration section 5 in the format shown in Figure 3. Ru.

Further, when image data is supplied from the image information reading section 1, the image data extraction section 6 extracts data to be registered as image information from the image data, and transfers this data to the image data compression section 7.
supply to.

When the image data compression section 7 receives the image data from the image data extraction section 6, it compresses it in a predetermined compression format and supplies it to the registration section 5.

When the registration section 5 is supplied with compressed image data from the image data compression section 7 and a code from the code conversion section 4, it associates these data and registers them on the optical disc 9.

Thus, in this example, filing 'iA
If you insert document 8 in @, the search target information described in this document 8 will be automatically extracted and this will be added to the document 8.
Since the documents 8 are registered on the optical disk 9 together with the image information, even when there are a large number of documents 8, they can be efficiently filed.

Furthermore, in the above-mentioned embodiment, the filing device according to the present invention was explained with reference to patent documents, but as shown in FIG. Extract this from document 1
It may also be filed on the optical disc 9 together with the image information of 0.

[Effects of the Invention] As explained above, according to the present invention, if the search title is standardized, necessary search target information etc. can be automatically input simply by reading the document. This allows input costs to be significantly reduced.

[Brief explanation of the drawing]

Fig. 1 is a block diagram showing an embodiment of a filing device according to the present invention, Fig. 2 is a schematic diagram showing an example of a document to be filed seven times in the same embodiment, and Fig. 3 is a schematic diagram showing an example of a document created by the same embodiment. A schematic diagram showing an example of search data, FIG. 4 is according to the present invention)? FIG. 5 is a block diagram showing an example of a conventional filing device. 1... Image information import unit 2... Search title recognition unit (key name recognition unit) 3...
・Information import unit (search information extraction unit) 5...Filing unit (registration unit) 8...Document agent Jl": i4 Kamizo Yoshii Stupid Man

Claims (1)

[Claims] (1) An image information reading section that reads an image of a document; a search title recognition section that recognizes a search title part from the image information obtained by this image information reading section; A filing device comprising: an information import section that imports information at a position corresponding to a retrieved title; and a filing section that files information retrieved by the information import section.