Method and apparatus for reading and recording text in digital form

Apparatus and a method of reading and recording in digital form human readable text, in which a plurality of sets of character recognition trees (6E,F,G) are stored in a data processing memory (6), pages of text are scanned by a document scanning device (1) and each line of text is segmented into a plurality of character patterns (3) each comprising a plurality of picture elements (4) characterised by the steps, for each line of text: (a) determining which set of recognition trees has the smallest reject ratio of unidentified characters to identified characters, (b) choosing that set and displaying (8) the line of text highlighting the unidentified character and (c) modifying the chosen set of character recognition trees in response to input signals correctly identifying the unidentified characters.