Novel Image coding and Inverse Halftoning

Texture is an important feature in images and has
been widely used in many applications. Based on the
classified textures, this book presents a novel
learning- and texture-based approach to design more
efficient image processing algorithms. For
context-based arithmetic coding, the block- and
texture-based training process is first applied to
train the multiple-template (MT) from the most
representative texture features. Based on the MT, we
next present a texture- and MT-based arithmetic
coding algorithm to compress error-diffused images.
For predictive coding, to improve the least
square approach, we present a texture-based training
process to construct the multiple-window (MW) for
various image contents. Based on the MW, the texture-
and MW-based prediction scheme is presented to
compress gray images. For inverse halftoning, based
on the proposed variance gain-based decision tree, a
texture-based training process is presented to
construct a lookup tree-table which will be used in
the reconstructing process. In the reconstructing
process, we propose an edge-based refinement scheme
to enhance the quality of the the
reconstructed gray image.