Document image compression for digital preservation, storage or transmission is an important research topic in recent years. Yet research into Chinese brush calligraphy classics digitization and their preservation with application to digital libraries remains in its infancy. In this paper, our research centers around Chinese brush calligraphy handwriting image compression. Extensive research is conducted and a viable scheme is proposed for document image compression with application to Chinese classics digitization and digital preservation. A comparative study of DjVu, JPEG2000 and the proposed method is performed. As the experimental results suggest, the proposed method outperforms DjVu and JPEG2000 in edge preserving and overall perceptual quality.