Cryptogram Decoding for OCR Using Numerization Strings
Authors:
- 1 September 2007
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE) in Ninth International Conference on Document Analysis and Recognition (ICDAR 2007)
- Vol. 1 (15205363) , 208-212
Abstract
OCR systems for printed documents typically require large numbers of font styles and character models to work well. When given an unseen font, performance degrades even in the absence of noise. In this paper, we perform OCR in an unsupervised fashion without using any character models by using a cryptogram decoding algorithm. We present results on real and artificial OCR data.Keywords
This publication has 6 references indexed in Scilit:
- Improving Recognition of Novel Input with Similarity2006
- Combining Local and Global Image Features for Object Class Recognition2005
- A statistical learning approach to document image analysis2005
- Substitution deciphering based on HMMs with applications to compressed document processingIEEE Transactions on Pattern Analysis and Machine Intelligence, 2002
- Efficient clustering of high-dimensional data sets with application to reference matching2000
- Techniques for automatically correcting words in textACM Computing Surveys, 1992