Degraded word recognition based on segmental signal-to-noise ratio weighting.
Hidefumi KobatakeYousuke MatsunooPublished in: ICASSP (1) (1994)
Keyphrases
- signal to noise ratio
- word recognition
- noise reduction
- handwritten word recognition
- noisy environments
- handwriting recognition
- word segmentation
- cursive script
- character recognition
- hidden markov models
- edge detection
- matched filter
- document analysis
- gaussian noise
- historical documents
- minimum mean square error
- automatic speech recognition
- bit error rate
- received signal
- wiener filter
- root mean square error
- motion blur
- spatial filtering
- image segmentation
- speech enhancement