AbdulWahab Kabani and Mahmoud R. El-Sakka, "Adaptive Weighted Neighbors Lossless Image Coding", International Conference on Image Analysis and Recognition, ICIAR'2015, LNCS 9164, pp. 129-138, Springer-Verlag Berlin Heidelberg, July 2015, Niagara Falls, Canada.

Abstract

Adaptive Weighted Neighbors Lossless Image Coding AWN is a symmetric lossless image compression algorithm. AWN makes two initial predictions, creates a weighted combination of the initial predictions before adjusting the prediction to end up with the final prediction. In order to achieve more compression, we encode the error in multiple bins depending on the expected error magnitude. Also, instead of encoding the signed error, the algorithm attempts to guess the sign and encodes the error magnitude and whether guessing the sign was successful or not.