Improving text-independent phonetic segmentation based on the Microcanonical Multiscale Formalism.
Vahid KhanaghaKhalid DaoudiOriol PontHussein M. YahiaPublished in: ICASSP (2011)
Keyphrases
- text detection
- multiscale
- image segmentation
- text lines
- image processing
- edge detection
- segmentation algorithm
- speech recognition
- texture segmentation
- affine invariant
- coarse to fine
- level set
- scale space
- document images
- shape representation
- text mining
- superpixels
- knowledge representation
- image analysis
- natural language
- scale selection
- multiscale segmentation
- spoken term detection