Entropy Coders Based on Binary Forword Classification for Image Compression

Entropy coders as a noiseless compression method are widely used as end-point compression for images so there have been many contributions to increase of entropy coder performance and to reduction of entropy coder complexity. In this paper, we propose some entropy coders based on binary forward classification (BFC), BFC requires overhead of classification but there is no change between the amount of input information and that of classified output information, which we prove this property in this paper. And using the proved property, we propose entropy coders which are Golomb-Rice coder after BFC (BFC+GR) and arithmetic coder with BFC (BFC+A). The proposed entropy decoders do not have further complexity from BFC. Simulation results also show better performance than other entropy coders which have similar complexity to proposed coders.