Recognizer characterisation for combining handwriting recognition results at word level

Combining results of multiple approaches to recognizing handwriting is becoming increasingly attractive as a means of improving recognition performance. Most of the reported work concerns combination of results at character level (i.e. letter, digit). This paper concentrates on combination of results at the word level. Two approaches are presented: word list merging and linear combination. Word list merging requires no knowledge about the individual recognizers. Linear combination is an attempt to exploit the information about characteristics of individual recognizers. This appears more complex than in the case of combination of results at the character level. Dealing with words is more complex than with individual letters. Recognition of words is influenced by more factors, which can independently affect the recognition process. Characterisation of recognizers, used for word level combination, is more complex and requires more than a simple consideration of recognition success and failure. Handwriting data characterisation is introduced. A number of handwriting characteristics are extracted and used to guide the combination process. The choice of characteristics is made in the context of recognition methods used. No attempt at general characterisation of handwriting is made. Relations between handwriting characteristics and recognition results are observed and used to obtain characteristics of individual recognizers. Results of the two combination methods are reported and compared with another frequently used method for results combination, the Borda count.