Lossless image compression based on a fuzzy-clustered prediction

This paper proposes a compression algorithm relying on a classified linear-regression prediction followed by context based modeling and arithmetic coding of the outcome residuals. Images are partitioned into blocks, e.g., 8/spl times/8 or 16/spl times/16, and a minimum mean square (MMSE) linear predictor is calculated for each block. Fuzzy clustering is utilized to reduce the number of such predictors. Given a preset number of classes, a Fuzzy-C-Means algorithm produces an initial guess of classified predictors to be fed to an iterative procedure which classifies pixel blocks simultaneously refining the associated predictors. All the predictors are transmitted along with the label of each block. Coding time is affordable thanks to fast convergence of the iterative algorithms. Decoding is always performed in real time. The compression scheme provides impressive performances, especially when applied to X-ray images.