Improving Structured Text Recognition with Regular Expression Biasing
This method is useful for domain-specific applications like recognizing driver license numbers or drug names, but it is incremental as it builds on existing recognition techniques.
The paper tackles the problem of recognizing structured text by using regular expressions for biasing, resulting in significantly improved accuracy for matching text with a small degradation on other text.
We study the problem of recognizing structured text, i.e. text that follows certain formats, and propose to improve the recognition accuracy of structured text by specifying regular expressions (regexes) for biasing. A biased recognizer recognizes text that matches the specified regexes with significantly improved accuracy, at the cost of a generally small degradation on other text. The biasing is realized by modeling regexes as a Weighted Finite-State Transducer (WFST) and injecting it into the decoder via dynamic replacement. A single hyperparameter controls the biasing strength. The method is useful for recognizing text lines with known formats or containing words from a domain vocabulary. Examples include driver license numbers, drug names in prescriptions, etc. We demonstrate the efficacy of regex biasing on datasets of printed and handwritten structured text and measures its side effects.