Adaptive combination of linear predictors for lossless image compression

Lossless image coding is an essential requirement for medical imaging appli cations. Lossless image compression techniques usually have two major compo nents: adaptive prediction and adaptive entropy coding. The paper is concer ned with adaptive prediction. Recently, several researchers have studied pr ediction schemes in which the final prediction is formed by a combination o f a group of subpredictors. The authors present an overview of this new typ e of prediction technique. They show that the basic principle of adaptive p redictor combination has been extensively studied and applied to many scien ce and engineering problems. They then describe their own combination schem e, which is based on the estimation of the local prediction error variance. Experimental results show that the compression performance of the algorith ms that employ this new type of predictor is consistently better than that of state-of-the-art algorithms.