Neural net vector quantizers for discrete HMM-based on-line handwritten whiteboard-note recognition.
Joachim SchenkGerhard RigollPublished in: ICPR (2008)
Keyphrases
- neural nets
- vector quantizer
- vector quantization
- character recognition
- handwritten characters
- numeral strings
- image coding
- feed forward
- word recognition
- neural network
- entropy constrained
- hidden markov models
- handwriting recognition
- counter propagation
- writer independent
- backpropagation neural networks
- back propagation
- artificial neural networks
- word spotting
- handwritten words
- feature extraction
- frequency domain
- image compression
- lattice vector quantization
- finite state
- reinforcement learning
- learning algorithm