Merge-Based Color Quantization and Context Tree Modeling for Compression of Color Quantized Images

A two-stage lossless compression method based on a binary tree representation of colors and on context-based arithmetic coding has been recently proposed. We propose two improvements to this method: merge-based color quantization instead of the original splitting strategy, and context tree modeling optimized for each layer separately. The proposed method achieves better compression performance and a better reproduction quality in the color progression.