STABILIZATION OF PROBABILITY ESTIMATES BY CONDITIONING ON PRIOR DECISIONS OF A GIVEN CONTEXT

摘要

A system and method involving a statistical conditioning technique that improves the codingefficiency in compression systems which have unstable statistical properties by conditioningthe probability estimate for a given model context on prior decisions for that context thusenlarging the conditioning decision set. Each model context is expanded into a set of twocontexts which are the actual coding contexts. For a given probability estimation modelcontext, the selection of a coding context index is done on the basis of the previous codingdecision (1 or 0) for that model context. Thus, if a model context is assigned an index A,the coding context would be assigned an index 2*A + D(A)', where D(A)' is the previousbinary decision for model context A; the decision of which coding context to use turns onwhether the model context was used most recently for coding a 1 or a 0.