Finding words in alphabet soup: Inference on freeform character recognition for historical scripts
| Autor: | Nicholas R. Howe, R. Manmatha, Shaolei Feng |
|---|---|
| Rok vydání: | 2009 |
| Předmět: |
Vocabulary
Boosting (machine learning) Computer science business.industry Speech recognition media_common.quotation_subject Inference Speech processing computer.software_genre Intelligent word recognition ComputingMethodologies_PATTERNRECOGNITION Artificial Intelligence Signal Processing Word recognition ComputingMethodologies_DOCUMENTANDTEXTPROCESSING Computer Vision and Pattern Recognition Artificial intelligence Hidden Markov model business Cursive computer Software Natural language processing media_common |
| Zdroj: | Pattern Recognition. 42:3338-3347 |
| ISSN: | 0031-3203 |
| DOI: | 10.1016/j.patcog.2009.01.012 |
| Popis: | This paper develops word recognition methods for historical handwritten cursive and printed documents. It employs a powerful segmentation-free letter detection method based upon joint boosting with histograms of gradients as features. Efficient inference on an ensemble of hidden Markov models can select the most probable sequence of candidate character detections to recognize complete words in ambiguous handwritten text, drawing on character n-gram and physical separation models. Experiments with two corpora of handwritten historic documents show that this approach recognizes known words more accurately than previous efforts, and can also recognize out-of-vocabulary words. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full Text from ScienceDirect