Experiments in the Contextual Recognition of Cursive Script
Roger W. Ehrich, Karin Koehler · IEEE Transactions on Computers · 1975
In this paper a structure for a contextual cursive script recognition system is presented. This system makes use of letter context to determine word length, letter segmentation, and character identity to achieve contextual recognition at the word level. The system consists of a character recognizer that presents a set of best alternatives for each character to a contextual postprocessor whose task it is to determine the correct word.