| 摘要 |
<p>An optical character recognition system which automatically reads handwritten characters and the like which do not have to be printed in a special format, utilizes a computer with memory into which digitized pixel information from a page containing characters to be recognized are entered. The characters are assigned fields and are processed and recognized one at a time. A bit map is formed in memory of the character in a matrix of rows and columns (x-y coordinates) representing the pixels of the pattern constituting the character. The bit map is thinned to form, in another area of memory, another bit map of pixels arranged also in a matrix of rows and columns in x-y coordinates, of the skeleton (along the center of the pattern of pixels) of the character bit map. The skeleton pixel matrix and the character bit map are encoded into a plurality of recognition tables each with a string of codes. The table derived from the skeleton pixel matrix describes, with connection codes, the internal structure, row by row in accordance with the relative position of pixels in adjacent rows. Recognition tables derived from the pattern bit map and from the skeleton pixel matrix describe the character in terms of the relative position of the pixels along the borders of the character and provide a plurality of recognition strings, one in each table, for the front and rearviews of the character and for the shape of the holes in the character which are opened from the top (as in the numeral four) or opened from the bottom (as in the numeral seven). The recognition table derived from the skeleton pixel matrix is generated by reducing the connection codes to single codes for each line which embrace and characterize generally, rather than specifically, the connection codes for that row. From the recognition tables, the characters are recognized by searching recognition files containing blocks of successions of lines of code corresponding selectively for the codes in the recognition tables. The recognition file is arranged in hierarchal order so that the blocks in the file which represent characters having the lowest level of recognition difficulty in the character set to be recognized are searched first, the next highest level next and so forth. Generally, but not exclusively, the search for each character in the set is headed by a search through a block in the recognition file which excludes that character so that recognition can proceed to the blocks for the next character in the group of blocks for the same difficulty of recognition level or to blocks for the next level of recognition difficulty. In this manner characters are recognized with a high degree of reliability and an indication of failure to recognize the character occurs rather than misrecognition.</p> |
