초록 ▼

This disclosure relates to a character recognition method and apparatus through which highly accurate character recognition is capable of being executed inexpensively and at high speed. Character recognition is raised in speed by executing segmentation of character images from an input image and cha

This disclosure relates to a character recognition method and apparatus through which highly accurate character recognition is capable of being executed inexpensively and at high speed. Character recognition is raised in speed by executing segmentation of character images from an input image and character recognition from the segmented character images in parallel by separate processors without use being made of a special communication processor. After the character images have been segmented from the input image, the results of segmentation are evaluated and the character images are segmented further based upon the results of evaluation. Methods of evaluation include a method involving finding a standard character size from characters written in character images segmented initially, and adopting the difference between the standard size and the size of each segmented character image, and a method involving quantifying results of character recognition of character images segmented initially, and evaluating the results of quantification.

대표청구항 ▼

A character recognition method comprising: an input step of inputting an image original; a segmentation step of segmenting character image blocks, which are written in said image original inputted by said input step; a recognition step of performing character recognition by comparing extracted featu

A character recognition method comprising: an input step of inputting an image original; a segmentation step of segmenting character image blocks, which are written in said image original inputted by said input step; a recognition step of performing character recognition by comparing extracted features of the segmented character image blocks with information that has been stored in a prescribed character dictionary, and obtaining a degree of similarity; a setting step of setting identification information indicating whether or not each of the segmented character image blocks requires resegmentation based on the degree of similarity; a generation step of generating, for a group of consecutive segmented character image blocks that require resegmentation based on the identification information, a plurality of new character image blocks by using all possible consecutive combinations of the segmented character image blocks in the group that begin with a first segmented character image block in the group, subject to the condition that each new character image block has a width which is within a predetermined value; a repetition control step of repeatedly returning to said recognition step so as to perform character recognition on the plurality of new character image blocks; and an output step of outputting a recognition result of the new character image block having the highest degree of similarity among the plurality of new character image blocks which have been generated in said generation step and subjected to character recognition in said recognition step, as a recognition result for a resegmented image block.