Lexicon-free handwritten word spotting using character HMMs

摘要

For retrieving keywords from scanned handwritten documents, we present a word spotting system that is based on character Hidden Markov Models. In an efficient lexicon-free approach, arbitrary keywords can be spotted without pre-segmenting text lines into words. For a multi-writer scenario on the IAM off-line database as well as for two single writer scenarios on historical data sets, it is shown that the proposed learning-based system outperforms a standard template matching method.