Introduction to Digital Typography
Introduction to Digital Typography
Document Image Decoding by Heuristic Search
IEEE Transactions on Pattern Analysis and Machine Intelligence
Supervised Template Estimation for Document Image Decoding
IEEE Transactions on Pattern Analysis and Machine Intelligence
Prototype Extraction and Adaptive OCR
IEEE Transactions on Pattern Analysis and Machine Intelligence
Optical Character Recognition: An Illustrated Guide to the Frontier
Optical Character Recognition: An Illustrated Guide to the Frontier
Document Image Decoding Using Markov Source Models
IEEE Transactions on Pattern Analysis and Machine Intelligence
Document Image Decoding Using Iterated Complete Path Search with Subsampled Heuristic Scoring
ICDAR '01 Proceedings of the Sixth International Conference on Document Analysis and Recognition
Pattern Classification (2nd Edition)
Pattern Classification (2nd Edition)
Robust document image understanding technologies
Proceedings of the 1st ACM workshop on Hardcopy document processing
Style Consistent Classification of Isogenous Patterns
IEEE Transactions on Pattern Analysis and Machine Intelligence
Adaptive OCR with Limited User Feedback
ICDAR '05 Proceedings of the Eighth International Conference on Document Analysis and Recognition
Document image analysis for digital libraries
Proceedings of the 2006 international workshop on Research issues in digital libraries
Hi-index | 0.00 |
We show that document image decoding (DID) supervisedtraining algorithms, as a result of recent refinements,achieve high accuracy with low manual effort even underconditions of severe image degradation in both trainingand test data. We describe improvements in DID trainingof character template, set-width, and channel (noise) models.Large-scale experimental trials, using synthetically degradedimages of text, have established two new and practicallyimportant advantages of DID algorithms:1. high accuracy ( 99% chraracters correct) in decodingusing models trained on even severely degradedimages from the same distribution; and2. greatly improved accuracy (