Document filing apparatus and method having document registration, comparison and retrieval capabilities

Claim 2

Document filing apparatus comprising: file means for storing a plurality of documents having line data, character data and image data; image input means for inputting image data of a new document; character-line extraction means for extracting line data and character data from the input image data; document retrieval means for retrieving from the file means one of the stored documents to be compared with the new document, the one of stored documents being similar to the new document; difference detection means for detecting differences of line data, character data, and image data between the new document and the retrieved document, said differences called difference data; and register means for registering the difference data and identification information of the retrieved document as the new document in said file means or for registering line data, character data and image data of the new document in said file means, if there is no stored document that compares to the new document.

Claim 3

Document filing apparatus according to claim 1, wherein the document retrieval means, upon determining that one of said stored documents is similar to the new document, retrieves the one of the stored documents.

Claim 4

Document filing apparatus according to claim 1, wherein each of the stored documents has associated therewith a predetermined mark representing a kind of form of the associated document, and the document retrieval means retrieves the stored document to be compared with the new document from the file means in accordance with the predetermined mark.

Claim 5

Document filing apparatus according to claim 1, wherein the image input means inputs a blank form whose line and character data are similar to those of a document to be retrieved in the file means.

Claim 6

Document filing apparatus according to claim 4, wherein the character-line extraction means extracts line data and character data from the blank form, and the document retrieval means retrieves documents from the file means according to the line data and the character data extracted from the blank form.

Claim 7

Document filing method comprising the steps of: storing in a file means a plurality of documents having line data, character data and image data; inputting a new document as image data; extracting line data and character data from the input image data of the new document; retrieving from the file means one of said stored documents to be compared with the new document, the one of said stored documents being similar to the new document; detecting differences of line data, character data, and image data between the new document and the retrieved document, the differences being called difference data; and registering in the file means, either the difference data and identification information of the retrieved document as the new document or registering line data, character data and image data of the new document if there is no stored document that is similar to the new document.

Claim 8

Document filing method comprising the steps of: storing in a file means a plurality of documents having line data, character data and image data; inputting a new document as image data; extracting line data and character data from the input image data of the new document; retrieving from the file means one of said stored documents to be compared with the new document, the one of said stored documents being similar to the new document; detecting differences of line data, character data, and image data between the new document and the retrieved document, the differences being called difference data; and registering in the file means, either the difference data and identification information of the retrieved document as the new document or registering line data, character data and image data of the new document if there is no stored document that is similar to the new document.

Claim 9

Document filing apparatus, comprising: file means for storing a plurality of documents having line data, character data and image data; image input means for inputting image data of a new document; character-line extraction means for extracting line data and character data from the input image data; difference detection means for calculating difference data between the new document and each document stored in said file means, and for determining the one of the stored documents for which the difference data is smallest; and register means for registering the difference data and identification information for the determined document as the new document in said file means.

Claim 10

Document filing apparatus, comprising: file means for storing a plurality of documents having line data, character data and image data; image input means for inputting image data of a new document; character-line extraction means for extracting line data and character data from the input image data; difference detection means for calculating difference data between the new document and each document stored in said file means, and for determining the one of the stored documents for which the difference data is smallest; and register means for registering the difference data and identification information for the determined document as the new document in said file means.

Claim 11

Document filing apparatus according to claim 7, wherein the register means registers the line data, character data, and image data of the new document when the smallest difference data is greater than a predetermined threshold.

Claim 12

Document filing method comprising the steps of: storing in file means a plurality of documents having line data, character data and image data in file means; inputting a new document as image data; extracting line data and character data from the input image data of the new document; calculating difference data between the new document and each document stored in said file means; determining the one of the stored documents for which the difference data is smallest; and registering in the file means the difference data and identification information for the determined document as the new document.

Claim 13

Document filing method comprising the steps of: storing in file means a plurality of documents having line data, character data and image data in file means; inputting a new document as image data; extracting line data and character data from the input image data of the new document; calculating difference data between the new document and each document stored in said file means; determining the one of the stored documents for which the difference data is smallest; and registering in the file means the difference data and identification information for the determined document as the new document.

Claim 14

Document filing apparatus, comprising: file means for storing image data of a plurality of documents; image input means for inputting imaga data of a new document; extraction means for extracting line data from the new document input image data and extracting ones of the plurality of rectangles included in the new document according to the line data; rectangle calculation means for calculating a coordinate value of a mid-point of each extracted rectangle, and for calculating a circumscribed rectangle which includes the plurality of rectangles; document retrieval means for retrieving image data of one of the stored documents from the file means, for calculating coordinate values of mid-points of a plurality of rectangles of the retrieved document, and for calculating a circumscribed rectangle which includes the plurality of the rectangles of the retrieved document; means for determining if a size of the circumscribed rectangle of the new document is the same as a size of the circumscribed of the retrieved document and, if the same size, for checking if the coordinate value of the mid-point of each rectangle of the new document overlaps the mid-point of each rectangle of the retrieved document; selection means for selecting rectangles of the retrieved document whose mid-point coordinate value coincides with one of the mid-point coordinate values of the new document; and output means for, if a selection ratio of a number of the selected rectangles to a total number of all rectangles of the retrieved document is above a threshold value, outputting the retrieved document as a retrieval result.

Claim 15

Document filing apparatus, comprising: file means for storing image data of a plurality of documents; image input means for inputting imaga data of a new document; extraction means for extracting line data from the new document input image data and extracting ones of the plurality of rectangles included in the new document according to the line data; rectangle calculation means for calculating a coordinate value of a mid-point of each extracted rectangle, and for calculating a circumscribed rectangle which includes the plurality of rectangles; document retrieval means for retrieving image data of one of the stored documents from the file means, for calculating coordinate values of mid-points of a plurality of rectangles of the retrieved document, and for calculating a circumscribed rectangle which includes the plurality of the rectangles of the retrieved document; means for determining if a size of the circumscribed rectangle of the new document is the same as a size of the circumscribed of the retrieved document and, if the same size, for checking if the coordinate value of the mid-point of each rectangle of the new document overlaps the mid-point of each rectangle of the retrieved document; selection means for selecting rectangles of the retrieved document whose mid-point coordinate value coincides with one of the mid-point coordinate values of the new document; and output means for, if a selection ratio of a number of the selected rectangles to a total number of all rectangles of the retrieved document is above a threshold value, outputting the retrieved document as a retrieval result.

Claim 16

Document filing apparatus according to claim 10, wherein the document retrieval means retrieves image data of another one of the stored documents from the file means, if the size of the circumscribed rectangle of the new document is not the same as the size of the circumscribed rectangle of the retrieved document or if the selection ratio is not above the threshold value.

Claim 17

Document filing apparatus according to claim 10, wherein the rectangle calculation means calculates the coordinate value of a mid-point of a unified rectangle formed from neighboring rectangles in the new document, and the checking means checks the coordinate value of the mid-point of the unified rectangle.

Claim 18

Document filing apparatus according to claim 10, wherein the selection means selects the unified rectangle if the coordinate value of the mid-point of the unified rectangle coincides with the coordinate value of the mid-point of one of the rectangles of the retrieved document.

Claim 19

Document filing method comprising the steps of: storing in file means image data of a plurality of documents; inputting image data of a new document; extracting line data from the input image data and a plurality of rectangles included in the new document according to the line data; retrieving image data of one of the stored documents from the file means; calculating a coordinate value of a mid-point of each extracted rectangle and a circumscribed rectangle which includes the plurality of rectangles; calculating coordinate values of mid-points of a plurality of rectangles of the retrieved document and calculating a circumscribed rectangle which includes the plurality of rectangles of the retrieved document; determining if a size of the circumscribed rectangle of the new document is the same as a size of the circumscribed rectangle of the retrieved document; checking, in case of the same size, if the coordinate value of the mid-point of each rectangle of the new document overlaps the mid-point of each rectangle of the retrieved document; selecting rectangles of the retrieved document whose mid-point coordinate value coincides with one of the mid-point coordinate values of the new document; and outputting the retrieved document as a retrieval result, if a ratio of a number of the selected rectangles to a total number of all rectangles of the retrieved document is above a threshold value.

Claim 20

Document filing method comprising the steps of: storing in file means image data of a plurality of documents; inputting image data of a new document; extracting line data from the input image data and a plurality of rectangles included in the new document according to the line data; retrieving image data of one of the stored documents from the file means; calculating a coordinate value of a mid-point of each extracted rectangle and a circumscribed rectangle which includes the plurality of rectangles; calculating coordinate values of mid-points of a plurality of rectangles of the retrieved document and calculating a circumscribed rectangle which includes the plurality of rectangles of the retrieved document; determining if a size of the circumscribed rectangle of the new document is the same as a size of the circumscribed rectangle of the retrieved document; checking, in case of the same size, if the coordinate value of the mid-point of each rectangle of the new document overlaps the mid-point of each rectangle of the retrieved document; selecting rectangles of the retrieved document whose mid-point coordinate value coincides with one of the mid-point coordinate values of the new document; and outputting the retrieved document as a retrieval result, if a ratio of a number of the selected rectangles to a total number of all rectangles of the retrieved document is above a threshold value.

Claim 21

A program storage device readable by a computer in a computer system, including a program of instructions executable by the computer to perform method steps, wherein said computer system includes a file means which stores a plurality of documents having line data, character data, and image data, said method steps comprising: extracting line data and character data from an input image data of a new document; retrieving from the file means one of said stored documents to be compared with the new document, the one of said stored documents being similar to the new document; detecting differences of line data, character data, and image data between the new document and the retrieved document, the differences called difference data; and registering in the file means either the difference data and identification information of the retrieved document as the new document or registering line data, character data, and image data of the new document if there is no stored document that is similar to the new document.

Claim 22

A program storage device readable by a computer in a computer system, including a program of instructions executable by the computer to perform method steps, wherein said computer system includes a file means which stores a plurality of documents having line data, character data, and image data, said method steps comprising: extracting line data and character data from an input image data of a new document; retrieving from the file means one of said stored documents to be compared with the new document, the one of said stored documents being similar to the new document; detecting differences of line data, character data, and image data between the new document and the retrieved document, the differences called difference data; and registering in the file means either the difference data and identification information of the retrieved document as the new document or registering line data, character data, and image data of the new document if there is no stored document that is similar to the new document.

Claim 23

A program storage device readable by a computer in a computer system, including a program of instructions executable by the computer to perform method steps, wherein said computer system includes a file means which stores a plurality of documents having line data, character data and image data, said method steps comprising; storing in file means a plurality of documents having line data, character data and image data in file means; inputting a new document as image data; extracting line data and character data from the input image data of the new document; calculating difference data between the new document and each document stored in said file means; determining the one of the stored documents for which the difference data is smallest; and registering in the file means the difference data and identification information for the determined document as the new document.

Claim 24

A program storage device readable by a computer in a computer system, including a program of instructions executable by the computer to perform method steps, wherein said computer system includes a file means which stores a plurality of documents having line data, character data and image data, said method steps comprising; storing in file means a plurality of documents having line data, character data and image data in file means; inputting a new document as image data; extracting line data and character data from the input image data of the new document; calculating difference data between the new document and each document stored in said file means; determining the one of the stored documents for which the difference data is smallest; and registering in the file means the difference data and identification information for the determined document as the new document.

Claim 25

A program storage device readable by a computer in a computer system, including a program of instructions executable by the computer to perform method steps, wherein said computer system includes a file means which stores a plurality of documents having line data, character data and image data, said method steps comprising; extracting line data from an input image data of a new document and a plurality of rectangles included in the new document according to the line data; retrieving image data of one of the stored documents from the file means; calculating a coordinate value of a mid-point of each extracted rectangle and a circumscribed rectangle which includes the plurality of rectangles; calculating coordinate values of mid-points of a plurality of rectangles of the retrieved document and calculating a circumscribed rectangle which includes the plurality of rectangles of the retrieved document; determining if a size of the circumscribed rectangle of the new document is the same as a size of the circumscribed rectangle of the retrieved document; checking, in case of the same size, if the coordinate value of the mid-point of each rectangle of the new document overlaps the mid-point of each rectangle of the retrieved document; selecting rectangles of the retrieved document whose mid-point coordinate value coincides with one of the mid-point coordinate values of the new document; and outputting the retrieved document as a retrieval result, if a ratio of a number of the selected rectangles to a total number of all rectangles of the retrieved document is above a threshold value.

Claim 26

A program storage device readable by a computer in a computer system, including a program of instructions executable by the computer to perform method steps, wherein said computer system includes a file means which stores a plurality of documents having line data, character data and image data, said method steps comprising; extracting line data from an input image data of a new document and a plurality of rectangles included in the new document according to the line data; retrieving image data of one of the stored documents from the file means; calculating a coordinate value of a mid-point of each extracted rectangle and a circumscribed rectangle which includes the plurality of rectangles; calculating coordinate values of mid-points of a plurality of rectangles of the retrieved document and calculating a circumscribed rectangle which includes the plurality of rectangles of the retrieved document; determining if a size of the circumscribed rectangle of the new document is the same as a size of the circumscribed rectangle of the retrieved document; checking, in case of the same size, if the coordinate value of the mid-point of each rectangle of the new document overlaps the mid-point of each rectangle of the retrieved document; selecting rectangles of the retrieved document whose mid-point coordinate value coincides with one of the mid-point coordinate values of the new document; and outputting the retrieved document as a retrieval result, if a ratio of a number of the selected rectangles to a total number of all rectangles of the retrieved document is above a threshold value.