Hitachi, Ltd.

Shozo Kadota

Yoshiji Fujimoto

Michio Yasuda

Shinichi Hayashi

Hiromichi Fujisawa

Toshihiro Hananoi

Yasuaki Nakano

Hitoshi Komatsu

Mitsunori Oka

Kiyomichi Kurino

Masao Yamamoto

Pattern segmentation and recognition in which hand-written characters are transformed electrically into 2-dimensional image patterns, wherein if ambiguity exists in segmenting a unit pattern including a character from the image patterns, character recognition is not made compulsively, but a plurality of possible unit patterns are first established. Then, the various unit patterns are segmented, and each unit pattern is identified to be a partial pattern, linked patterns, etc., so that each character is recognized on a basis of total judgement, whereby ambiguity of segmentation is resolved.

System and method for segmentation and recognition of patterns

Disclosed is a device for detecting a displacement between patterns comprising a picking-up device, first and second memory means for storing a first and a second patterns picked up by said picking up device, means for shifting said first and second patterns relative to each other thereby obtaining a correlation value for each shift between them, means for obtaining a maximum correlation value and deriving a displacement in terms of the number of shifts which gives said maximum correlation value.

Device for detecting displacement between patterns

A pattern segmenting system for a pattern recognizing device, including a pattern memory for storing character information, means for reading out the character information from the pattern memory, means for detecting the start and end points of a subpattern in the character information and for calculating the center of the subpattern in accordance with the start and end points thereof, means for deciding whether or not the center thereof is within a selected character frame, and means for segmenting the subpattern when the center thereof is within the selected character frame and for regarding the frame as a blank when it is not within the selected character frame.

Pattern segmenting system for a pattern recognizing device

In order to recognize confusable characters exactly, a pattern recognition system is described wherein, when more than two reference patterns having a similarity to the input unknown pattern larger than a predetermined threshold level are detected by a first discriminating circuit, one or more suitable feature comparators are selected from a plurality of feature comparators by the control signals corresponding to the reference patterns or the category thereof detected by the first discriminating circuit and the selected one or more feature comparators determine the identity of the input unknown pattern by comparing selected features thereof to the corresponding portions of the plural reference patterns under consideration.

Pattern recognition system

Character recognition apparatus which converts two dimensional feature information derived from an unknown character into one dimensional feature information, such as a code train, and recognizes the unknown character by comparing the converted one dimensional feature information with corresponding information relating to selected reference characters. The starting point of the one dimensional feature information representing the unknown character is selected by scanning the terminal points of the unknown character in a direction predetermined for each particular reference character as indicated by a scan parameter stored with the one dimensional feature information of the reference character. The starting point is then selected on the basis of a relationship including the scan parameter of the reference character and the feature points of the unknown character.

Character recognition apparatus

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

Hachioji, Japan

Shozo Kadota