Predictor blending technique for lossless and near-lossless image coding
Grzegorz Ulacha, Ryszard Stasiński
Abstract
Grzegorz Ulacha, Ryszard Stasiński
Abstract
In the paper a lossless image coding method intended for real-time hardware intra-frame video coding system is described. Data modeling stage of the method is based on predictor blending idea, the following entropy coding stage is realized as an advanced adaptive context arithmetic coder. The stages can be separated by a quantizer, in which case the coder is becoming a near-lossless one. Experiments show that indeed, the new algorithm is not only time efficient, but has also excellent data compression performance.
OpenAlex reports 7 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
In the paper a lossless image coding method intended for real-time hardware intra-frame video coding system is described. Data modeling stage of the method is based on predictor blending idea, the following entropy coding stage is realized as an advanced adaptive context arithmetic coder. The stages can be separated by a quantizer, in which case the coder is becoming a near-lossless one. Experiments show that indeed, the new algorithm is not only time efficient, but has also excellent data compression performance.
Key concepts: Lossless compression, Entropy encoding, Context-adaptive binary arithmetic coding, Adaptive coding, Data compression, Context-adaptive variable-length coding, Computer science, Tunstall coding