Hitachi, Ltd.

Toshihiro Hananoi

Hiromichi Fujisawa

Yasuaki Nakano

Atsushi Hatakeyama

Junichi Higashino

Shigeru Kakumoto

Yoshiji Fujimoto

Michio Yasuda

Shinichi Hayashi

Tatsuya Murakami

Hidefumi Masuzaki

Masaaki Fujinawa

Haruo Takeda

Satoshi Ito

Shozo Kadota

Shunichi Yajima

Sachiko Kawachiya

Minoru Oota

Hirohide Endou

A document storage and retrieval system is provided with means for storing a document body in the form of image, means for storing text information in the form of a character code string for retrieval, means for executing a retrieval with reference to the text information, and means for displaying a document image relating thereto on a retrieval terminal according to the retrieval result. Such a form of the system is available for retrieving the full contents of a document and also for displaying the document body printed in a format easy to read straight in the form of image. Accordingly, users are capable of retrieving documents with arbitrary words and also capable of reading even such a document as is complicated to include mathematical expressions and charts through a terminal in the form of image, the same as on paper. Further, the invention provides a system wherein the text information for retrieval is extracted automatically from the document image through character recognition. Since a precision of the character recognition has not been satisfactory hitherto, a visual retrieval and correction have been carried out without fail by operators. However, there is no necessity for the operators to attend therefor according to the invention. Thus, the text information for retrieval can be generated at the cost of practical time and money even in case of volumes of documents.

Document storage and retrieval system for storing and retrieving

When image information of documents is stored in the image storage of an image filing device, the title information is entered for each document for easy retrieval of the image information after filing. For entry of the title information for each document into the image filing device, it is necessary to fill in title information in the title entry format, supply the title information entry format through the image input unit of the image filing device and to recognize the characters of the title information. This title information entry format is automatically designed and entered into the image storage by the operator who supplies the title information and file name by operating the keyboard of the image filing device. When the operator designates one of the plural title information entry formats entered for each file, the title information entry format thus designated is printed out by the image output unit of the image filing device. The printed title information entry format is filled-in with the title information and set in the image input unit of the image filing device along with a document to be filed. With the above operations, the image filing device recognizes the title information and files it along with the document into the image storage.

Method for generating title information entry format and method for

A document storage and retrieval system stores a document body in the form of an image, storing text information in the form of a character code string for retrieval, and executing a retrieval with reference to the text information, followed by displaying a document image relating thereto on a retrieval terminal according to the retrieval result. Such a form of the system is available for retrieving the full contents of a document and also for displaying the document body printed in a format easy to read straight in the form of an image.

Document storage and retrieval system

A document image to be corrected is read out from an image file which stores document images, converted into dots, and printed on a correction paper. An operator writes correction information on the correction paper which is in turn inputted to the image file system with an image reader. The image file system erases the dot image information of the image inputted with the image reader. The obtained correction information is synthesized with the original image read out from the image file. The synthesized corrected image is stored in the image file.

Image data filing system and image data correcting method

A binary-coding which is effected converts a video signal into binary-coded outputs corresponding to white and black. More particularly, the binary-coding method and apparatus produces a black peak signal by following the black peak of a video signal, and makes a determination based on this black peak signal and video signal as to whether or not the video signal is rich in the black components so as to provide a binary slice select reference signal which takes the H level when the video signal is rich in the black components, holds the slice select reference signal for several scans in the vertical direction and for several bits in the horizontal direction, selects a binary-coded waveform obtained by slicing the video signal at a predetermined slice level when at least one among the slice select reference signals held is found at the level H indicating that the video signal is rich in black components, and outputs it as binary-coded output. Accordingly, a binary-coding method and apparatus is provided wherein when any hand-written or printed dark characters or the like are detected as a video signal and are binary-coded into white and black levels, the binary-coded output can be reproduced without being smudged or fuzzed dark.

Video signal binary-coding apparatus and method

For halftone digital image data, an edge portion of the image is extracted for each pixel, and based on a density of each pixel or an average density of each group of a plurality of adjacent or neighboring pixels, there are extracted pixels of a background other than the edge portion of the image. Thereafter, the image data is subdivided into a plurality of blocks each including a predetermined number of pixels. Based on distribution states of the pixels of the block judged to belong to the edge and those of the block judged to belong to the background, a domain recognition is conducted to determine whether the block is a binary block or a halftone block. In addition, for each block, a state of areas in blocks encircling the block is examined such that depending on the state, an expansion/contraction processing is repeatedly achieved a predetermined times for the halftone domain, thereby separating the image data into a binary domain and a halftone domain in a real-time fashion.

Document image entry system

In a pattern recognition device for recognizing an unknown pattern in accordance with the magnitude of the similarities between the unknown pattern and a plurality of standard patterns, the similarity between the unknown pattern and one of the standard patterns is detected as follows. Similarities are detected at first in respective shifting conditions where the unknown and standard patterns are relatively shifted from each other over the first limited extent, including the condition without the shift. The maximum value of these similarities is then detected. The similarities are further detected in respective shifting conditions where the unknown and standard patterns are relatively shifted from each other over the second extend larger than the first limited extent, when the shifting condition which gave the maximum value is that without relative shift.

Method and device for detecting the similarity between standard and

A pattern processing system comprising a filtering processor which enhances the edge of a character signal picked up by an image pickup tube, a comparator which binary-codes the character signal on the basis of threshold values of two different levels, a circuit which provides the logical product between outputs from the filtering processor and the comparator, and a filtering processor which subjects the output of the circuit to a filtering processing by the use of a smoothing filter.

Pattern processing system

A pattern preliminary processing system wherein high and low frequency noise other than the pattern signal from a pick-up tube are eliminated by a filtering circuit thereby to 'relieve' the pattern against the background and the thus obtained pattern signal is tri-valued into signals of high, medium and low levels by means of a processing circuit having two different threshold values, and then the signal of the medium level is converted into a signal of high or low level with reference to peripheral information thereby to finally bi-value the pattern signal.

Pattern preliminary processing system

A character pattern recognition system is disclosed in which an unknown input pattern of character and a standard pattern of character are described and maintained in the form of a sequence of Freeman direction codes. One-to-multiple and multiple-to-one correspondence is permitted between the codes, to obtain a specific correspondence between them in which the sum of the difference between direction codes is minimum, so that the minimum value of code difference is used as a measure of determining the similarity between the two patterns.

Growing community of inventors

Kanagawa, Japan

Toshihiro Hananoi