Name
Abstract | ||
|---|---|---|
We present a semi-Markov model for recognizing scene text that integrates character and word segmen- tation with recognition. Using wavelet features, it re- quires only approximate location of the text baseline and font size; no binarization or prior word segmen- tation is necessary. Our system is aided by a lexicon, yet it also allows non-lexicon words. To facilitate in- ference with a large lexicon, we use an approximate Viterbi beam search. Our system performs robustly on low-resolution images of signs containing text in fonts atypical of documents. |
Year | DOI | Venue |
|---|---|---|
2008 | 10.1109/ICPR.2008.4761818 | Tampa, FL |
Keywords | Field | DocType |
Markov processes,character recognition,document handling,feature extraction,image segmentation,maximum likelihood estimation,text analysis,wavelet transforms,approximate Viterbi beam search,character segmentation,discriminative semiMarkov model,nonlexicon words,robust scene text recognition,scene text recognition,wavelet features,word segmentation | Pattern recognition,Computer science,Markov model,Feature extraction,Speech recognition,Text segmentation,Image segmentation,Lexicon,Artificial intelligence,Hidden Markov model,Discriminative model,Viterbi algorithm | Conference |
ISSN | ISBN | Citations |
1051-4651 E-ISBN : 978-1-4244-2175-6 | 978-1-4244-2175-6 | 7 |
PageRank | References | Authors |
0.79 | 11 | 3 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Jerod J. Weinman | 1 | 72 | 5.80 |
| Erik G. Miller | 2 | 1861 | 126.56 |
| A. Hanson | 3 | 1348 | 304.11 |