Quadtree and statistical model-based lossless binary image compression method

Abstract

This paper presents a new lossless binary image compression method. The method consists of four stages: preprocessing, quadtree compressing, run length compressing and statistical model-based compressing. The preprocessing stage is to reduce the entropy of a binary image. In the quadtree compressing stage, a breadth first traversal (BFT) linear quadtree is used to compress the image. The run length compressing stage uses the run length method to encode the tree list and colour list of the BFT linear quadtree. The statistical model-based compressing stage applies the Huffman coding algorithm to compress the remaining data of the BFT linear quadtree further. Experimental results show that this method can provide an impressive compression result.