Title :
Partially hidden Markov models
Author :
Forchhammer, Soren ; Rissanen, Jorma
Author_Institution :
Inst. of Telecommun., Tech. Univ., Lyngby, Denmark
fDate :
7/1/1996 12:00:00 AM
Abstract :
Partially hidden Markov models (PHMM) are introduced. They differ from the ordinary HMMs in that both the transition probabilities of the hidden states and the output probabilities are conditioned on past observations. As an illustration they are applied to black and white image compression where the hidden variables may be interpreted as representing noncausal pixels
Keywords :
data compression; hidden Markov models; image coding; image representation; probability; black and white image compression; hidden states; noncausal pixels; output probabilities; partially hidden Markov models; past observations; transition probabilities; Councils; Data compression; Hidden Markov models; Image coding; Parameter estimation; Pixel; Probability distribution; Speech recognition; State-space methods; Text recognition;
Journal_Title :
Information Theory, IEEE Transactions on
