N-gram language models for document image decoding
Gary E. Kopec, Maya R. Said, Kris Popat · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · 2001
This paper explores the problem of incorporating linguistic constraints into document image decoding, a communication theory approach to document recognition. Probabilistic character n-grams (n=2--5) are used in a two-pass strategy where the decoder first uses a very weak language model to generate a lattice of candidate output strings. These are then re-scored in the second pass using the full language model. Experimental results based on both synthesized and scanned data show that this approach is capable of improving the error rate by a factor of two to ten depending on the quality of the data and the details of the language model used.