On the estimation of the probability distribution of a non stationary source for lossless data compression
Jonas D. Pfefferman, Hector J. Gonzalez, Bruno Cernuschi-Frías · 2002
In this work, a method for the estimation of the probabilities of a non stationary source is proposed in order to adapt quickly to the source statistics changes. This estimation is based on the time intervals between occurrences of each symbol. The method is applied to lossless compression of images.