Method of recognizing characters

A method of recognizing characters of headers and characters of data on a document, comprising the steps of:

• extracting character strings on the document by reading the document;
• distinguishing between headers and data on the document by determining the positional relationship between the character strings;
• determining character attributes of the data by recognizing characters of the character strings of the headers using a header recognition dictionary;
• and
• recognizing characters of the character strings of the data according to the determined character attributes of the data,
• wherein said step of distinguishing between headers and data comprises the steps of:
• extracting character strings belonging to one logical row from the positional relationship between the character strings;
• extracting logical rows which have the same arrangement of character strings, as a block;
• extracting character strings which have the same structure as a logical column in each of blocks;
• determining an associated relationship between logical columns from the positional relationship between the logical columns;
• and
• determining headers of the logical columns from the positional relationship between the blocks.

A recognition method of a layout of a document comprising:

• a procedure of extracting character strings having the characters of the headers of the data and the characters of the data on the document from an image of the document;
• a procedure of distinguishing between the headers of data and the data on the document by detecting each position of said extracted character strings, categorizing said extracted character strings into rows and columns according to said detected position of said extracted character strings, categorizing said extracted character strings into associated blocks according to said categorized rows and columns, detecting positions of each of said blocks, and determining the header from the detected positional relationship between said blocks;
• and
• a procedure of determining the layout from said distinguished headers of data and the data on the document.

A storage medium storing a program for recognition of a layout of a document comprising:

• a first program for extracting character strings having the characters of the headers of the data and the characters of the data on the document from an image of the document;
• a second program for distinguishing between the headers of data and the data on the document by detecting each position of said extracted character strings, categorizing said extracted character strings into rows and columns according to said detected position of said extracted character strings, categorizing said extracted character strings into associated blocks according to said categorized rows and columns, detecting positions of each of said blocks, and determining the header from the detected positional relationship between said blocks;
• and
• a third program for determining the layout from said distinguished headers of the data and the data on the documment.